{"generated":"2026-10-24T23:23:29Z","window_days":30,"status":{"hf_listed":10813,"hf_candidates":218,"prescreen_drops":{"older fine-tune <100 likes":852,"merge/abliterated variant":242,"older, below threshold, never shown":632,"test/demo/checkpoint name":47,"fresh, no human signal":726,"re-upload of a trusted-org model name":160,"spam author":374},"spam_authors":["Allinsama","ArchiveStudio","AutomatosX","CollectionStudio","DanielMarius","Hebrish","Mihara-bot","MoritzMo123","Muapi","Thox-ai","anonymous-asr-review","att","chameleon-ai","fjzzq2002","mdagosta","mflux-community","myflow-ai","praiselab-picuslab","vosldtgbj","zhangjiaxin2012"],"hf_models":218,"github_fail":[],"github_ok":17,"ollama":260,"x_posts":965,"threshold":{"min_hf_likes":50,"min_hf_downloads":5000,"min_x_likes":100,"fresh_hours":72,"fresh_min_likes":2,"fresh_min_downloads":0,"fresh_max":60,"kept":185,"just_landed":54,"just_landed_unvetted":40,"main":131,"filtered":23,"trusted_kept":99,"junk_dropped":{"no weight files":5,"tiny/test repo":4,"empty card":1}},"real_quant_sizes":118,"archive":237,"fit_from_real_files":118,"fit_estimate_only":31},"models":[{"type":"model","id":"kindlingai/glm-5.3-exlr8-k2-k3-k4","name":"glm-5.3-exlr8-k2-k3-k4","org":"kindlingai","date":"2026-10-08","license":"glm-5.3","params":null,"params_src":null,"active":null,"ctx":1048576,"kv_bytes_tok":89856,"dtypes":[],"pipeline":"text-generation","modality":"text","summary_what":"EXLR8 is a new quantization format.","arch":{"class":"GlmMoeDsaForCausalLM","type":"MoE","experts":256,"experts_active":8},"claims":["GLM-5.3's top-8 routing turns small numeric differences into different experts: two BF16 runs that differ only in their batching score mean 0.11 (median 0.004) against each other on 32,768 positions of the held-out text."],"has_safetensors":true,"has_weights":true,"card_len":19797,"used_storage":837527438131,"created":"2026-10-08T03:20:20Z","first_seen":"2026-10-22T23:27:46Z","fresh":true,"url":"https://huggingface.co/kindlingai/glm-5.3-exlr8-k2-k3-k4","likes":22,"downloads":935966,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"935966 downloads","velocity":2316.5,"lane":"fresh","vetted":true,"quant_sizes":{},"summary":"EXLR8 is a new quantization format. Mixture-of-experts (256 experts, 8 per token). Context: 1M tokens (config). License: glm-5.3. Vendor-reported: \"GLM-5.3's top-8 routing turns small numeric differences into different experts: two BF16 runs that differ only in their batching score mean 0.11 (median 0.004) against each other on 32,768 positions of the held-out text.\"","ollama":"https://ollama.com/library/glm-5.3","est":null,"license_class":"restricted"},{"type":"model","id":"MiniMaxAI/MiniMax-M3.1-Flash","name":"MiniMax-M3.1-Flash","org":"MiniMaxAI","date":"2026-10-17","license":"minimax-community","params":220490854016,"params_src":"safetensors","active":23000000000.0,"ctx":1048576,"kv_bytes_tok":122880,"dtypes":["F32","BF16","F8_E4M3","U8"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"MiniMax-M3.1-Flash is a native multimodal mixture-of-experts model for coding, agentic workflows, computer use, and cybersecurity.","arch":{"class":"MiniMaxM3SparseForConditionalGeneration","type":"MoE","experts":128,"experts_active":4,"active":23000000000.0,"active_src":"model card"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19802,"used_storage":252429151663,"created":"2026-10-17T06:37:50Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/MiniMaxAI/MiniMax-M3.1-Flash","likes":307,"downloads":5789,"finetune":false,"gated":false,"quants":[{"id":"mlx-community/MiniMax-M3.1-Flash-Text-3bit","kind":"MLX"},{"id":"mlx-community/MiniMax-M3.1-Flash-Text-4bit","kind":"MLX"},{"id":"mlx-community/MiniMax-M3.1-Flash-oQ3.5e-mtp","kind":"MLX"},{"id":"mlx-community/MiniMax-M3.1-Flash-oQ3e-mtp","kind":"MLX"},{"id":"mlx-community/MiniMax-M3.1-Flash-oQ4e-mtp","kind":"MLX"},{"id":"unsloth/MiniMax-M3.1-Flash","kind":"Unsloth"},{"id":"unsloth/MiniMax-M3.1-Flash-GGUF","kind":"GGUF"}],"x_mentions":[{"url":"https://x.com/TeksEdge/status/2112208149684105324","user":"TeksEdge","likes":12},{"url":"https://x.com/SarangMahatwo/status/2112367014119383229","user":"SarangMahatwo","likes":2},{"url":"https://x.com/Shivamshahi77/status/2114054313526075526","user":"Shivamshahi77","likes":0}],"x_likes":1104,"x_posts":68,"signal":"trusted org","velocity":712.99,"lane":"fresh","vetted":true,"quant_sizes":{"Q8":{"gb":259.1,"file":"UD-Q8_K_XL","repo":"unsloth/MiniMax-M3.1-Flash-GGUF","kind":"GGUF","community":false}},"summary":"MiniMax-M3.1-Flash is a native multimodal mixture-of-experts model for coding, agentic workflows, computer use, and cybersecurity. Mixture-of-experts, 220B total, 23B active (128 experts, 4 per token). Context: 1M tokens (config). License: minimax-community.","ollama":"https://ollama.com/library/minimax-m3.1-flash","est":{"weights":{"Q4":132.3,"Q8":259.1,"FP16":441.0},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 220.5B params"},"Q8":{"kind":"real","file":"UD-Q8_K_XL","repo":"unsloth/MiniMax-M3.1-Flash-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 220.5B params"}},"kv_gb":1.0,"kv_src":"config","need":{"Q4":134.3,"Q8":261.1,"FP16":443.0},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":"Q4","amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":"Q4","mac256":"Q4","mac512":"Q8","cpu32":null,"cpu64":null,"cpu128":null,"cpu256":"Q4"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":"estimate","amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":"estimate","mac256":"estimate","mac512":"real","cpu32":null,"cpu64":null,"cpu128":null,"cpu256":"estimate"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"kindlingai/glm-5.3-flash-exlr8-k2-k3-k4","name":"glm-5.3-flash-exlr8-k2-k3-k4","org":"kindlingai","date":"2026-10-14","license":"mit","params":null,"params_src":null,"active":null,"ctx":1048576,"kv_bytes_tok":46080,"dtypes":[],"pipeline":"image-text-to-text","modality":"vision","summary_what":"GLM-5.3-Flash is a MoE model: 45 layers (0-2 dense, 3-44 MoE), 288 routed experts with top-8 routing plus one shared expert per MoE layer, an MTP layer 45 and a vision tower.","arch":{"class":"Glm5NextForConditionalGeneration","type":"MoE","experts":288,"experts_active":8},"claims":["GLM-5.3-Flash is a MoE model: 45 layers (0-2 dense, 3-44 MoE), 288 routed experts with top-8 routing plus one shared expert per MoE layer, an MTP layer 45 and a vision tower."],"has_safetensors":true,"has_weights":true,"card_len":11689,"used_storage":714925712995,"created":"2026-10-14T13:59:14Z","first_seen":"2026-10-22T23:27:46Z","fresh":true,"url":"https://huggingface.co/kindlingai/glm-5.3-flash-exlr8-k2-k3-k4","likes":3,"downloads":18051,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"18051 downloads","velocity":72.39,"lane":"fresh","vetted":true,"quant_sizes":{},"summary":"GLM-5.3-Flash is a MoE model: 45 layers (0-2 dense, 3-44 MoE), 288 routed experts with top-8 routing plus one shared expert per MoE layer, an MTP layer 45 and a vision tower. Mixture-of-experts (288 experts, 8 per token). Context: 1M tokens (config). License: mit. Vendor-reported: \"GLM-5.3-Flash is a MoE model: 45 layers (0-2 dense, 3-44 MoE), 288 routed experts with top-8 routing plus one shared expert per MoE layer, an MTP layer 45 and a vision tower.\"","ollama":"https://ollama.com/library/glm-5.3-flash","est":null,"license_class":"permissive"},{"type":"model","id":"LiconStudio/Minimax-H3-VBVR","name":"Minimax-H3-VBVR","org":"LiconStudio","date":"2026-10-23","license":"apache-2.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"Licon VBVR is an image-to-video LoRA for MiniMax-H3, trained on VBVR data rebuilt and reprocessed specifically for H3.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":2914,"used_storage":188960551,"created":"2026-10-23T10:38:40Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/LiconStudio/Minimax-H3-VBVR","likes":76,"downloads":939,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"76 likes","velocity":27.62,"lane":"fresh","vetted":true,"quant_sizes":{},"summary":"Licon VBVR is an image-to-video LoRA for MiniMax-H3, trained on VBVR data rebuilt and reprocessed specifically for H3. License: apache-2.0.","est":null,"license_class":"permissive"},{"type":"model","id":"tencent/WeVisOne-VL-4B","name":"WeVisOne-VL-4B","org":"tencent","date":"2026-10-22","license":"apache-2.0","params":4830581248,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":147456,"dtypes":["F32"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"WeVisOne-VL (WeChat Vision All-in-One) is a unified vision-language model developed by WeChat Vision, Tencent Inc.","arch":{"class":"Qwen3VLForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":2687,"used_storage":19372555374,"created":"2026-10-22T09:48:14Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/tencent/WeVisOne-VL-4B","likes":22,"downloads":524,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":8.87,"lane":"fresh","vetted":true,"quant_sizes":{"Q4":{"gb":2.7,"file":"Q4_K_M","repo":"mradermacher/WeVisOne-VL-4B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":5.2,"file":"Q8_0","repo":"mradermacher/WeVisOne-VL-4B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":9.7,"file":"F16","repo":"mradermacher/WeVisOne-VL-4B-GGUF","kind":"GGUF","community":false}},"summary":"WeVisOne-VL (WeChat Vision All-in-One) is a unified vision-language model developed by WeChat Vision, Tencent Inc. Dense, 4.8B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":2.7,"Q8":5.2,"FP16":9.7},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/WeVisOne-VL-4B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/WeVisOne-VL-4B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"mradermacher/WeVisOne-VL-4B-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.2,"kv_src":"config","need":{"Q4":4.9,"Q8":7.4,"FP16":11.9},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"HuggingFaceBio/Carbon-X","name":"Carbon-X","org":"HuggingFaceBio","date":"2026-10-21","license":"apache-2.0","params":14822046720,"params_src":"safetensors","active":null,"ctx":40960,"kv_bytes_tok":163840,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"Carbon-X is a 14.8B-parameter decoder-only language model trained jointly on DNA sequences and natural-language text.","arch":{"class":"Qwen3ForCausalLM","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19794,"used_storage":29656327254,"created":"2026-10-21T10:21:04Z","first_seen":"2026-10-22T11:23:11Z","fresh":true,"url":"https://huggingface.co/HuggingFaceBio/Carbon-X","likes":145,"downloads":570,"finetune":true,"gated":false,"quants":[{"id":"HuggingFaceBio/Carbon-X-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"145 likes","velocity":8.41,"lane":"fresh","vetted":true,"quant_sizes":{"Q4":{"gb":9.6,"file":"Q4_K_M","repo":"HuggingFaceBio/Carbon-X-GGUF","kind":"GGUF","community":false},"Q8":{"gb":15.8,"file":"Q8_0","repo":"HuggingFaceBio/Carbon-X-GGUF","kind":"GGUF","community":false},"FP16":{"gb":29.7,"file":"BF16","repo":"HuggingFaceBio/Carbon-X-GGUF","kind":"GGUF","community":false}},"summary":"Carbon-X is a 14.8B-parameter decoder-only language model trained jointly on DNA sequences and natural-language text. Dense, 15B params. Context: 40K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":9.6,"Q8":15.8,"FP16":29.7},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"HuggingFaceBio/Carbon-X-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"HuggingFaceBio/Carbon-X-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"HuggingFaceBio/Carbon-X-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.3,"kv_src":"config","need":{"Q4":11.9,"Q8":18.1,"FP16":32.0},"fit":{"nv8":null,"gpu12":null,"gpu16":"Q4","nv24":"Q8","nv32":"Q8","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q4","amd24":"Q8","strix":"FP16","gpu24":"Q8","gpu32":"Q8","mac16":"Q4","mac24":"Q4","mac36":"Q8","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q8","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"RBDash-Team/Xing-VL1.0-Decision","name":"Xing-VL1.0-Decision","org":"RBDash-Team","date":"2026-10-22","license":"apache-2.0","params":26085330160,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"Xing-VL1.0-Decision is a 27B multimodal typed decision model.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":6336,"used_storage":52193402988,"created":"2026-10-22T02:37:18Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/RBDash-Team/Xing-VL1.0-Decision","likes":3,"downloads":492,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":7.2,"lane":"fresh","vetted":false,"quant_sizes":{"Q4":{"gb":16.5,"file":"Q4_K_M","repo":"mradermacher/Xing-VL1.0-Decision-GGUF","kind":"GGUF","community":false},"Q8":{"gb":29.2,"file":"Q8_0","repo":"mradermacher/Xing-VL1.0-Decision-GGUF","kind":"GGUF","community":false},"FP16":{"gb":52.2,"file":"official BF16 safetensors","repo":"RBDash-Team/Xing-VL1.0-Decision","kind":"safetensors","community":false}},"summary":"Xing-VL1.0-Decision is a 27B multimodal typed decision model. Dense, 26B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":16.5,"Q8":29.2,"FP16":52.2},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/Xing-VL1.0-Decision-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/Xing-VL1.0-Decision-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"RBDash-Team/Xing-VL1.0-Decision","fmt":"safetensors","community":false}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":19.6,"Q8":32.3,"FP16":55.3},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"tencent/WeVisOne-VL-2B","name":"WeVisOne-VL-2B","org":"tencent","date":"2026-10-22","license":"apache-2.0","params":2441744384,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":114688,"dtypes":["F32"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"WeVisOne-VL (WeChat Vision All-in-One) is a unified vision-language model developed by WeChat Vision, Tencent Inc.","arch":{"class":"Qwen3VLForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":2687,"used_storage":9816802246,"created":"2026-10-22T09:47:50Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/tencent/WeVisOne-VL-2B","likes":25,"downloads":393,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":6.79,"lane":"fresh","vetted":true,"quant_sizes":{"Q4":{"gb":1.3,"file":"Q4_K_M","repo":"mradermacher/WeVisOne-VL-2B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":2.6,"file":"Q8_0","repo":"mradermacher/WeVisOne-VL-2B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":4.9,"file":"F16","repo":"mradermacher/WeVisOne-VL-2B-GGUF","kind":"GGUF","community":false}},"summary":"WeVisOne-VL (WeChat Vision All-in-One) is a unified vision-language model developed by WeChat Vision, Tencent Inc. Dense, 2.4B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":1.3,"Q8":2.6,"FP16":4.9},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/WeVisOne-VL-2B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/WeVisOne-VL-2B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"mradermacher/WeVisOne-VL-2B-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.9,"kv_src":"config","need":{"Q4":3.2,"Q8":4.5,"FP16":6.8},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"sky7350/Mica-v0.5-4B","name":"Mica-v0.5-4B","org":"sky7350","date":"2026-09-27","license":"apache-2.0","params":4659865088,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":131072,"dtypes":["BF16","F32"],"pipeline":"text-classification","modality":"text","summary_what":"Mica is a 4B model that answers your questions about an input with a probability for every answer you allow.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19693,"used_storage":29657241757,"created":"2026-09-27T20:51:18Z","first_seen":"2026-10-22T23:27:46Z","fresh":true,"url":"https://huggingface.co/sky7350/Mica-v0.5-4B","likes":15,"downloads":2224,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":3.44,"lane":"fresh","vetted":false,"quant_sizes":{"Q4":{"gb":2.8,"file":"Q4_K_M","repo":"mradermacher/Mica-v0.5-4B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":5.0,"file":"Q8_0","repo":"mradermacher/Mica-v0.5-4B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":9.3,"file":"F16","repo":"mradermacher/Mica-v0.5-4B-GGUF","kind":"GGUF","community":false}},"summary":"Mica is a 4B model that answers your questions about an input with a probability for every answer you allow. Dense, 4.7B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":2.8,"Q8":5.0,"FP16":9.3},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/Mica-v0.5-4B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/Mica-v0.5-4B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"mradermacher/Mica-v0.5-4B-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":4.9,"Q8":7.1,"FP16":11.4},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"bottlecapai/ThinkingCap-Qwen3.8-Flash-Next","name":"ThinkingCap-Qwen3.8-Flash-Next","org":"bottlecapai","date":"2026-10-22","license":"qwen-community-license-1.0","params":179999981459,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":98304,"dtypes":["BF16","I64"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"Thinking is on at reasoningeffort=xhigh, the chat template's own default, with the base model's recommended sampling &mdash; temperature=1.0, topp=0.95, topk=20, minp=0.0 &mdash; used unchanged for Ours.","arch":{"class":"Qwen4ExpForConditionalGeneration","type":"MoE","experts":512,"experts_active":10},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19746,"used_storage":360160625472,"created":"2026-10-22T15:15:16Z","first_seen":"2026-10-22T23:27:46Z","fresh":true,"url":"https://huggingface.co/bottlecapai/ThinkingCap-Qwen3.8-Flash-Next","likes":29,"downloads":163,"finetune":true,"gated":false,"quants":[{"id":"bottlecapai/ThinkingCap-Qwen3.8-Flash-Next-FP8","kind":"quant"},{"id":"bottlecapai/ThinkingCap-Qwen3.8-Flash-Next-GGUF","kind":"GGUF"},{"id":"bottlecapai/ThinkingCap-Qwen3.8-Flash-Next-NVFP4","kind":"quant"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":3.42,"lane":"fresh","vetted":false,"quant_sizes":{"Q4":{"gb":113.2,"file":"Q4_K_M","repo":"bottlecapai/ThinkingCap-Qwen3.8-Flash-Next-GGUF","kind":"GGUF","community":false},"Q8":{"gb":191.0,"file":"Q8_0","repo":"bottlecapai/ThinkingCap-Qwen3.8-Flash-Next-GGUF","kind":"GGUF","community":false},"FP16":{"gb":359.2,"file":"BF16","repo":"bottlecapai/ThinkingCap-Qwen3.8-Flash-Next-GGUF","kind":"GGUF","community":false}},"summary":"Thinking is on at reasoningeffort=xhigh, the chat template's own default, with the base model's recommended sampling &mdash; temperature=1.0, topp=0.95, topk=20, minp=0.0 &mdash; used unchanged for Ours. Mixture-of-experts, 180B total (512 experts, 10 per token). Context: 256K tokens (config). License: qwen-community-license-1.0.","ollama":"https://ollama.com/library/qwen3.8-flash-next","est":{"weights":{"Q4":113.2,"Q8":191.0,"FP16":359.2},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"bottlecapai/ThinkingCap-Qwen3.8-Flash-Next-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"bottlecapai/ThinkingCap-Qwen3.8-Flash-Next-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"bottlecapai/ThinkingCap-Qwen3.8-Flash-Next-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.8,"kv_src":"config","need":{"Q4":115.0,"Q8":192.8,"FP16":361.0},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":"Q4","spark2":"Q8","amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":"Q4","mac256":"Q4","mac512":"FP16","cpu32":null,"cpu64":null,"cpu128":"Q4","cpu256":"Q8"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":"real","mac256":"real","mac512":"real","cpu32":null,"cpu64":null,"cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"restricted"},{"type":"model","id":"MinimaLabs/KLM-1.5","name":"KLM-1.5","org":"MinimaLabs","date":"2026-10-22","license":"apache-2.0","params":98975308,"params_src":"safetensors","active":null,"ctx":2048,"kv_bytes_tok":24576,"dtypes":["I64","F32","I32"],"pipeline":"text-generation","modality":"text","summary_what":"KLM 1.5 is a 99M-parameter base language model.","arch":{"class":"KLMForCausalLM","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":6923,"used_storage":396104400,"created":"2026-10-22T12:21:22Z","first_seen":"2026-10-22T23:27:46Z","fresh":true,"url":"https://huggingface.co/MinimaLabs/KLM-1.5","likes":15,"downloads":171,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":3.15,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"KLM 1.5 is a 99M-parameter base language model. Dense, 99M params. Context: 2K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":0.1,"Q8":0.1,"FP16":0.2},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.1B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.1B params"}},"kv_gb":0.2,"kv_src":"config","need":{"Q4":1.3,"Q8":1.3,"FP16":1.4},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16","name":"NVIDIA-Nemotron-3.5-Bolt-E4B-BF16","org":"nvidia","date":"2026-10-05","license":"openmdw-1.1","params":7996156490,"params_src":"safetensors","active":null,"ctx":65536,"kv_bytes_tok":86016,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"NVIDIA-Nemotron-3.5-Bolt-E4B-BF16 is a small language model (SLM) further post-trained from Gemma-4-E4B-it by NVIDIA.","arch":{"class":"Gemma4ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19486,"used_storage":16055022874,"created":"2026-10-05T04:25:48Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16","likes":44,"downloads":1453,"finetune":true,"gated":false,"quants":[{"id":"ggml-org/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF","kind":"GGUF"},{"id":"mlx-community/NVIDIA-Nemotron-3.5-Bolt-E4B-OptiQ-4bit","kind":"MLX"},{"id":"nvidia/NVIDIA-Nemotron-3.5-Bolt-E4B-BF16-DFlash-NVFP4","kind":"quant"},{"id":"unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":3.15,"lane":"fresh","vetted":true,"quant_sizes":{"Q4":{"gb":5.1,"file":"UD-Q4_K_XL","repo":"unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":8.2,"file":"UD-Q8_K_XL","repo":"unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":15.1,"file":"BF16","repo":"unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF","kind":"GGUF","community":false}},"summary":"NVIDIA-Nemotron-3.5-Bolt-E4B-BF16 is a small language model (SLM) further post-trained from Gemma-4-E4B-it by NVIDIA. Dense, 8B params. Context: 64K tokens (config). License: openmdw-1.1.","ollama":"https://ollama.com/library/nemotron-3.5-bolt","est":{"weights":{"Q4":5.1,"Q8":8.2,"FP16":15.1},"weights_src":{"Q4":{"kind":"real","file":"UD-Q4_K_XL","repo":"unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"UD-Q8_K_XL","repo":"unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"unsloth/NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":6.8,"Q8":9.9,"FP16":16.8},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"inclusionAI/Ling-3.0-flash-Sante","name":"Ling-3.0-flash-Sante","org":"inclusionAI","date":"2026-10-19","license":"mit","params":127486405600,"params_src":"safetensors","active":5100000000.0,"ctx":262144,"kv_bytes_tok":48384,"dtypes":["F32","BF16"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"BailingMoeV3ForCausalLM","type":"MoE","experts":512,"experts_active":8,"active":5100000000.0,"active_src":"model card"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":8186,"used_storage":254993298756,"created":"2026-10-19T17:30:04Z","first_seen":"2026-10-22T23:27:46Z","fresh":true,"url":"https://huggingface.co/inclusionAI/Ling-3.0-flash-Sante","likes":47,"downloads":324,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":2.95,"lane":"fresh","vetted":true,"quant_sizes":{"Q8":{"gb":136.4,"file":"Q8_0","repo":"AtomicChat/Ling-3.0-flash-Sante-GGUF","kind":"GGUF","community":true},"FP16":{"gb":255.0,"file":"official BF16 safetensors","repo":"inclusionAI/Ling-3.0-flash-Sante","kind":"safetensors","community":false}},"summary":"Text generation model from inclusionAI (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 127B total, 5.1B active (512 experts, 8 per token). Context: 256K tokens (config). License: mit.","est":{"weights":{"Q4":76.5,"Q8":136.4,"FP16":255.0},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 127.5B params"},"Q8":{"kind":"real","file":"Q8_0","repo":"AtomicChat/Ling-3.0-flash-Sante-GGUF","fmt":"GGUF","community":true},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"inclusionAI/Ling-3.0-flash-Sante","fmt":"safetensors","community":false}},"kv_gb":0.4,"kv_src":"config","need":{"Q4":77.9,"Q8":137.8,"FP16":256.4},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":"Q4","spark1":"Q4","spark2":"Q8","amd16":null,"amd24":null,"strix":"Q4","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":"Q4","mac192":"Q8","mac256":"Q8","mac512":"FP16","cpu32":null,"cpu64":null,"cpu128":"Q4","cpu256":"Q8"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":"estimate","spark1":"estimate","spark2":"real","amd16":null,"amd24":null,"strix":"estimate","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":"estimate","mac192":"real","mac256":"real","mac512":"real","cpu32":null,"cpu64":null,"cpu128":"estimate","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"fastino/GLiDE","name":"GLiDE","org":"fastino","date":"2026-10-19","license":"fastino-community","params":26085330160,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"GLiDE is a decision model from Fastino.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":17708,"used_storage":52193622897,"created":"2026-10-19T10:25:20Z","first_seen":"2026-10-22T23:27:46Z","fresh":true,"url":"https://huggingface.co/fastino/GLiDE","likes":44,"downloads":343,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":2.91,"lane":"fresh","vetted":false,"quant_sizes":{"FP16":{"gb":52.2,"file":"official BF16 safetensors","repo":"fastino/GLiDE","kind":"safetensors","community":false}},"summary":"GLiDE is a decision model from Fastino. Dense, 26B params. Context: 256K tokens (config). License: fastino-community.","est":{"weights":{"Q4":15.7,"Q8":27.7,"FP16":52.2},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 26.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 26.1B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"fastino/GLiDE","fmt":"safetensors","community":false}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":18.8,"Q8":30.8,"FP16":55.3},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q8","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":"Q8","mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"estimate","strix":"real","gpu24":null,"gpu32":"estimate","mac16":null,"mac24":null,"mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"ltzheng/Qwen3.5-9B-General-Game-Stage1-3Shift-Shuffled","name":"Qwen3.5-9B-General-Game-Stage1-3Shift-Shuffled","org":"ltzheng","date":"2026-10-22","license":"apache-2.0","params":9653104368,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":131072,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":3811,"used_storage":849497360302,"created":"2026-10-22T03:24:21Z","first_seen":"2026-10-22T11:23:11Z","fresh":true,"url":"https://huggingface.co/ltzheng/Qwen3.5-9B-General-Game-Stage1-3Shift-Shuffled","likes":2,"downloads":180,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":2.68,"lane":"fresh","vetted":false,"quant_sizes":{"FP16":{"gb":19.3,"file":"official BF16 safetensors","repo":"ltzheng/Qwen3.5-9B-General-Game-Stage1-3Shift-Shuffled","kind":"safetensors","community":false}},"summary":"Image text to text model from ltzheng (HF pipeline tag; the card has no one-line description). Dense, 9.7B params. Context: 256K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/qwen3.5","est":{"weights":{"Q4":5.8,"Q8":10.3,"FP16":19.3},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 9.7B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 9.7B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"ltzheng/Qwen3.5-9B-General-Game-Stage1-3Shift-Shuffled","fmt":"safetensors","community":false}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":7.9,"Q8":12.4,"FP16":21.4},"fit":{"nv8":null,"gpu12":"Q4","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q4","mac24":"Q8","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":"estimate","gpu16":"estimate","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"estimate","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"estimate","mac24":"estimate","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"darkc0de/XORTRON-AgentRICO","name":"XORTRON-AgentRICO","org":"darkc0de","date":"2026-10-19","license":"","params":27781427952,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"XORTRON is an experimental research system.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":8616,"used_storage":55582996119,"created":"2026-10-19T23:11:46Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/darkc0de/XORTRON-AgentRICO","likes":4,"downloads":297,"finetune":true,"gated":false,"quants":[{"id":"bartowski/darkc0de_XORTRON-AgentRICO-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":2.5,"lane":"fresh","vetted":false,"quant_sizes":{"Q4":{"gb":16.8,"file":"Q4_K_M","repo":"darkc0de/XORTRON-AgentRICO-GGUF","kind":"GGUF","community":false},"Q8":{"gb":29.7,"file":"Q8_0","repo":"darkc0de/XORTRON-AgentRICO-GGUF","kind":"GGUF","community":false},"FP16":{"gb":54.7,"file":"BF16","repo":"bartowski/darkc0de_XORTRON-AgentRICO-GGUF","kind":"GGUF","community":false}},"summary":"XORTRON is an experimental research system. Dense, 28B params. Context: 256K tokens (config).","est":{"weights":{"Q4":16.8,"Q8":29.7,"FP16":54.7},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"darkc0de/XORTRON-AgentRICO-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"darkc0de/XORTRON-AgentRICO-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"bartowski/darkc0de_XORTRON-AgentRICO-GGUF","fmt":"GGUF","community":false}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":19.9,"Q8":32.8,"FP16":57.8},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"unknown"},{"type":"model","id":"som-doe2007/geoai-outdoor-assistant","name":"geoai-outdoor-assistant","org":"som-doe2007","date":"2026-10-23","license":"apache-2.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{},"claims":[],"has_safetensors":false,"has_weights":true,"card_len":6709,"used_storage":1118832126,"created":"2026-10-23T05:57:13Z","first_seen":"2026-10-22T23:27:46Z","fresh":true,"url":"https://huggingface.co/som-doe2007/geoai-outdoor-assistant","likes":3,"downloads":100,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":2.49,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"Text generation model from som-doe2007 (HF pipeline tag; the card has no one-line description). License: apache-2.0.","est":null,"license_class":"permissive"},{"type":"model","id":"bottlecapai/ThinkingCap-gemma-4-26B-A4B-it","name":"ThinkingCap-gemma-4-26B-A4B-it","org":"bottlecapai","date":"2026-10-23","license":"apache-2.0","params":25805936206,"params_src":"safetensors","active":4000000000.0,"ctx":262144,"kv_bytes_tok":245760,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"","arch":{"class":"Gemma4ForConditionalGeneration","type":"MoE","experts":128,"experts_active":8,"active":4000000000.0,"active_src":"name"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19706,"used_storage":51754409974,"created":"2026-10-23T09:55:44Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/bottlecapai/ThinkingCap-gemma-4-26B-A4B-it","likes":10,"downloads":70,"finetune":true,"gated":false,"quants":[{"id":"bartowski/bottlecapai_ThinkingCap-gemma-4-26B-A4B-it-GGUF","kind":"GGUF"},{"id":"bottlecapai/ThinkingCap-gemma-4-26B-A4B-it-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":2.14,"lane":"fresh","vetted":false,"quant_sizes":{"Q4":{"gb":17.0,"file":"Q4_K_M","repo":"bottlecapai/ThinkingCap-gemma-4-26B-A4B-it-GGUF","kind":"GGUF","community":false},"Q8":{"gb":26.9,"file":"Q8_0","repo":"bottlecapai/ThinkingCap-gemma-4-26B-A4B-it-GGUF","kind":"GGUF","community":false},"FP16":{"gb":50.5,"file":"BF16","repo":"bartowski/bottlecapai_ThinkingCap-gemma-4-26B-A4B-it-GGUF","kind":"GGUF","community":false}},"summary":"Image text to text model from bottlecapai (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 26B total, 4B active (128 experts, 8 per token). Context: 256K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/gemma4","est":{"weights":{"Q4":17.0,"Q8":26.9,"FP16":50.5},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"bottlecapai/ThinkingCap-gemma-4-26B-A4B-it-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"bottlecapai/ThinkingCap-gemma-4-26B-A4B-it-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"bartowski/bottlecapai_ThinkingCap-gemma-4-26B-A4B-it-GGUF","fmt":"GGUF","community":false}},"kv_gb":2.0,"kv_src":"config","need":{"Q4":20.0,"Q8":29.9,"FP16":53.5},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q8","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":"Q8","mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":"real","mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"NorviTech/armiger-ultra","name":"armiger-ultra","org":"NorviTech","date":"2026-10-18","license":"apache-2.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"text-classification","modality":"text","summary_what":"Armiger Ultra is a 9-head byte-level RoPE Transformer (d=256, 6 layers, 8 heads) with a 6-plugin pipeline for raw BitTorrent/DHT metadata (name, file-tree paths + sizes, optional tracker/DHT counters).","arch":{},"claims":[],"has_safetensors":false,"has_weights":true,"card_len":8601,"used_storage":442617492,"created":"2026-10-18T14:02:08Z","first_seen":"2026-10-22T23:27:46Z","fresh":true,"url":"https://huggingface.co/NorviTech/armiger-ultra","likes":10,"downloads":269,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":1.82,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"Armiger Ultra is a 9-head byte-level RoPE Transformer (d=256, 6 layers, 8 heads) with a 6-plugin pipeline for raw BitTorrent/DHT metadata (name, file-tree paths + sizes, optional tracker/DHT counters). License: apache-2.0.","est":null,"license_class":"permissive"},{"type":"model","id":"Hikari07jp/DeepSeek-V4.1-1017-Ablitrated-testmodel","name":"DeepSeek-V4.1-1017-Ablitrated-testmodel","org":"Hikari07jp","date":"2026-10-23","license":"mit","params":763205315794,"params_src":"safetensors","active":null,"ctx":1048576,"kv_bytes_tok":81920,"dtypes":["BF16","F32","F8_E4M3","I8"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"","arch":{"class":"DeepseekV41ForCausalLM","type":"MoE","experts":384,"experts_active":6},"claims":["Hidden size 5120, vocab 129,280, 384 routed experts (top-6) + 1 shared expert, MoE intermediate 2304."],"has_safetensors":true,"has_weights":true,"card_len":13362,"used_storage":510296920807,"created":"2026-10-23T06:27:17Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/Hikari07jp/DeepSeek-V4.1-1017-Ablitrated-testmodel","likes":6,"downloads":59,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":1.59,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"Image text to text model from Hikari07jp (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 763B total (384 experts, 6 per token). Context: 1M tokens (config). License: mit. Vendor-reported: \"Hidden size 5120, vocab 129,280, 384 routed experts (top-6) + 1 shared expert, MoE intermediate 2304.\"","est":{"weights":{"Q4":457.9,"Q8":810.9,"FP16":1526.4},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 763.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 763.2B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 763.2B params"}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":459.6,"Q8":812.6,"FP16":1528.1},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":null,"amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":null,"mac512":null,"cpu32":null,"cpu64":null,"cpu128":null,"cpu256":null},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":null,"amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":null,"mac512":null,"cpu32":null,"cpu64":null,"cpu128":null,"cpu256":null},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"tattrongvu/Jeffan-31b-1710","name":"Jeffan-31b-1710","org":"tattrongvu","date":"2026-10-19","license":"apache-2.0","params":31273086512,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":983040,"dtypes":["BF16"],"pipeline":"text-classification","modality":"text","summary_what":"","arch":{"class":"Gemma4ForConditionalGeneration","type":"dense"},"claims":["The Decision Index runs use the engine in inference/, which also reads a question a second time (the input shown twice) when its top-1 probability is below 0.9, and codes up to 255 options."],"has_safetensors":true,"has_weights":true,"card_len":10284,"used_storage":62578507890,"created":"2026-10-19T07:06:45Z","first_seen":"2026-10-22T11:23:11Z","fresh":true,"url":"https://huggingface.co/tattrongvu/Jeffan-31b-1710","likes":9,"downloads":199,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":1.53,"lane":"fresh","vetted":false,"quant_sizes":{"Q4":{"gb":18.7,"file":"Q4_K_M","repo":"mradermacher/Jeffan-31b-1710-GGUF","kind":"GGUF","community":false},"Q8":{"gb":33.4,"file":"Q8_0","repo":"mradermacher/Jeffan-31b-1710-GGUF","kind":"GGUF","community":false},"FP16":{"gb":62.5,"file":"official BF16 safetensors","repo":"tattrongvu/Jeffan-31b-1710","kind":"safetensors","community":false}},"summary":"Text classification model from tattrongvu (HF pipeline tag; the card has no one-line description). Dense, 31B params. Context: 256K tokens (config). License: apache-2.0. Vendor-reported: \"The Decision Index runs use the engine in inference/, which also reads a question a second time (the input shown twice) when its top-1 probability is below 0.9, and codes up to 255 options.\"","est":{"weights":{"Q4":18.7,"Q8":33.4,"FP16":62.5},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/Jeffan-31b-1710-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/Jeffan-31b-1710-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"tattrongvu/Jeffan-31b-1710","fmt":"safetensors","community":false}},"kv_gb":8.1,"kv_src":"config","need":{"Q4":27.8,"Q8":42.5,"FP16":71.6},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":null,"strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":"Q4","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":null,"cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":null,"cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"Valen-Team/Valen-0.8B","name":"Valen-0.8B","org":"Valen-Team","date":"2026-10-05","license":"apache-2.0","params":854305641,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":49152,"dtypes":["F32"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"","arch":{"class":"ValenQwenForDecisionMaking","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":13091,"used_storage":5265677971,"created":"2026-10-05T20:40:35Z","first_seen":"2026-10-24T11:24:09Z","fresh":true,"url":"https://huggingface.co/Valen-Team/Valen-0.8B","likes":2,"downloads":676,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":1.48,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"Image text to text model from Valen-Team (HF pipeline tag; the card has no one-line description). Dense, 854M params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":0.5,"Q8":0.9,"FP16":1.7},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.9B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.9B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.9B params"}},"kv_gb":0.4,"kv_src":"config","need":{"Q4":1.9,"Q8":2.3,"FP16":3.1},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"nvidia/InternVideo2-CLIP-anomaly-detection","name":"InternVideo2-CLIP-anomaly-detection","org":"nvidia","date":"2026-10-22","license":"nvidia-internal-scientific-research-and-development-model-license","params":372752640,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"","modality":"","summary_what":"Input Parameters: Video is a temporal sequence of two-dimensional RGB frames; text is a one-dimensional token sequence.","arch":{},"claims":["The fine-tune improves the measured domain retrieval score (+8.78 percentage points mAP over the base) while reducing Kinetics-400 Top-1 accuracy by 0.92 percentage points versus the recorded base reference."],"has_safetensors":true,"has_weights":true,"card_len":19605,"used_storage":8216736746,"created":"2026-10-22T18:23:57Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/nvidia/InternVideo2-CLIP-anomaly-detection","likes":4,"downloads":59,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":1.19,"lane":"fresh","vetted":true,"quant_sizes":{},"summary":"Input Parameters: Video is a temporal sequence of two-dimensional RGB frames; text is a one-dimensional token sequence. 373M params. License: nvidia-internal-scientific-research-and-development-model-license. Vendor-reported: \"The fine-tune improves the measured domain retrieval score (+8.78 percentage points mAP over the base) while reducing Kinetics-400 Top-1 accuracy by 0.92 percentage points versus the recorded base reference.\"","est":{"weights":{"Q4":0.2,"Q8":0.4,"FP16":0.7},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.4B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.4B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.4B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.7,"Q8":1.9,"FP16":2.2},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"ufakai/ufakzeka-kulak","name":"ufakzeka-kulak","org":"ufakai","date":"2026-10-22","license":"apache-2.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"automatic-speech-recognition","modality":"audio","summary_what":"ufakzeka-kulak (\"kulak\" is Turkish for ear) is a small Turkish speech-to-text model by ufak AI, trained from scratch and released under Apache-2.0.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":18816,"used_storage":1002238026,"created":"2026-10-22T23:36:32Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/ufakai/ufakzeka-kulak","likes":2,"downloads":54,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":1.17,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"ufakzeka-kulak (\"kulak\" is Turkish for ear) is a small Turkish speech-to-text model by ufak AI, trained from scratch and released under Apache-2.0. License: apache-2.0.","est":null,"license_class":"permissive"},{"type":"model","id":"kakaocorp/kanana-2-vision-encoder-medium","name":"kanana-2-vision-encoder-medium","org":"kakaocorp","date":"2026-10-12","license":"kanana-open-license","params":426886912,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":114688,"dtypes":["BF16"],"pipeline":"image-feature-extraction","modality":"other","summary_what":"Kanana-2-vision-encoder is Kakao's vision encoder series for vision-language models.","arch":{"class":"Kanana2VisionEncoder","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19739,"used_storage":853919704,"created":"2026-10-12T11:49:21Z","first_seen":"2026-10-22T11:23:11Z","fresh":true,"url":"https://huggingface.co/kakaocorp/kanana-2-vision-encoder-medium","likes":11,"downloads":318,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":1.1,"lane":"fresh","vetted":false,"quant_sizes":{"FP16":{"gb":0.9,"file":"official BF16 safetensors","repo":"kakaocorp/kanana-2-vision-encoder-medium","kind":"safetensors","community":false}},"summary":"Kanana-2-vision-encoder is Kakao's vision encoder series for vision-language models. Dense, 427M params. License: kanana-open-license.","est":{"weights":{"Q4":0.3,"Q8":0.5,"FP16":0.9},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.4B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.4B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"kakaocorp/kanana-2-vision-encoder-medium","fmt":"safetensors","community":false}},"kv_gb":0.9,"kv_src":"config","need":{"Q4":2.2,"Q8":2.4,"FP16":2.8},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"internlm/Atria-Dawn","name":"Atria-Dawn","org":"internlm","date":"2026-10-23","license":"","params":753329940480,"params_src":"safetensors","active":null,"ctx":1048576,"kv_bytes_tok":89856,"dtypes":["BF16","F32"],"pipeline":"","modality":"","summary_what":"Atria Dawn is a new-generation agentic model developed by the Shanghai Artificial Intelligence Laboratory.","arch":{"class":"GlmMoeDsaForCausalLM","type":"MoE","experts":256,"experts_active":8},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":20000,"used_storage":1506690922047,"created":"2026-10-23T03:21:40Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/internlm/Atria-Dawn","likes":6,"downloads":41,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":1.07,"lane":"fresh","vetted":true,"quant_sizes":{"FP16":{"gb":1506.7,"file":"official BF16 safetensors","repo":"internlm/Atria-Dawn","kind":"safetensors","community":false}},"summary":"Atria Dawn is a new-generation agentic model developed by the Shanghai Artificial Intelligence Laboratory. Mixture-of-experts, 753B total (256 experts, 8 per token). Context: 1M tokens (config).","est":{"weights":{"Q4":452.0,"Q8":800.4,"FP16":1506.7},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 753.3B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 753.3B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"internlm/Atria-Dawn","fmt":"safetensors","community":false}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":453.7,"Q8":802.1,"FP16":1508.4},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":null,"amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":null,"mac512":null,"cpu32":null,"cpu64":null,"cpu128":null,"cpu256":null},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":null,"amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":null,"mac512":null,"cpu32":null,"cpu64":null,"cpu128":null,"cpu256":null},"basis":"mixed: real files + estimate"},"license_class":"unknown"},{"type":"model","id":"nvidia/GR00T-N1.7-AppleToPlate-cosine40k-checkpoint-15000","name":"GR00T-N1.7-AppleToPlate-cosine40k-checkpoint-15000","org":"nvidia","date":"2026-10-13","license":"","params":3144016000,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"robotics","modality":"robotics","summary_what":"","arch":{"class":"Gr00tN1d7"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":9026,"used_storage":25540817621,"created":"2026-10-13T20:23:12Z","first_seen":"2026-10-22T23:27:46Z","fresh":true,"url":"https://huggingface.co/nvidia/GR00T-N1.7-AppleToPlate-cosine40k-checkpoint-15000","likes":0,"downloads":279,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":1.04,"lane":"fresh","vetted":true,"quant_sizes":{},"summary":"Robotics model from nvidia (HF pipeline tag; the card has no one-line description). 3.1B params.","est":{"weights":{"Q4":1.9,"Q8":3.3,"FP16":6.3},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 3.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 3.1B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 3.1B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":3.4,"Q8":4.8,"FP16":7.8},"fit":{"nv8":"Q8","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"unknown"},{"type":"model","id":"ccortezb/FinOptix-14B-v2","name":"FinOptix-14B-v2","org":"ccortezb","date":"2026-10-23","license":"apache-2.0","params":14770033664,"params_src":"safetensors","active":null,"ctx":32768,"kv_bytes_tok":196608,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"FinOptix-14B v2 is an open-weights FinOps specialist fine-tuned on Qwen 2.5 14B Instruct.","arch":{"class":"Qwen2ForCausalLM","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":5016,"used_storage":29551556126,"created":"2026-10-23T01:50:48Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/ccortezb/FinOptix-14B-v2","likes":2,"downloads":43,"finetune":true,"gated":false,"quants":[{"id":"ccortezb/FinOptix-14B-v2-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.99,"lane":"fresh","vetted":false,"quant_sizes":{"Q4":{"gb":9.0,"file":"Q4_K_M","repo":"ccortezb/FinOptix-14B-v2-GGUF","kind":"GGUF","community":false},"Q8":{"gb":15.7,"file":"Q8_0","repo":"mradermacher/FinOptix-14B-v2-GGUF","kind":"GGUF","community":false},"FP16":{"gb":29.5,"file":"official BF16 safetensors","repo":"ccortezb/FinOptix-14B-v2","kind":"safetensors","community":false}},"summary":"FinOptix-14B v2 is an open-weights FinOps specialist fine-tuned on Qwen 2.5 14B Instruct. Dense, 15B params. Context: 32K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":9.0,"Q8":15.7,"FP16":29.5},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"ccortezb/FinOptix-14B-v2-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/FinOptix-14B-v2-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"ccortezb/FinOptix-14B-v2","fmt":"safetensors","community":false}},"kv_gb":1.6,"kv_src":"config","need":{"Q4":11.6,"Q8":18.3,"FP16":32.1},"fit":{"nv8":null,"gpu12":null,"gpu16":"Q4","nv24":"Q8","nv32":"Q8","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q4","amd24":"Q8","strix":"FP16","gpu24":"Q8","gpu32":"Q8","mac16":"Q4","mac24":"Q4","mac36":"Q8","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q8","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"GrootTeam/Groot-19B-A3B-261023","name":"Groot-19B-A3B-261023","org":"GrootTeam","date":"2026-10-23","license":"apache-2.0","params":19432294256,"params_src":"safetensors","active":3000000000.0,"ctx":262144,"kv_bytes_tok":81920,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"Groot is a family of open-weight reasoning models for complex real-world tasks.","arch":{"class":"Qwen3_5MoeForConditionalGeneration","type":"MoE","experts":128,"experts_active":8,"active":3000000000.0,"active_src":"name"},"claims":["Groot-19B-A3B-261023 beats both Qwen3.6-35B-A3B and Gemma-4-26B-A4B on 12 of the 16 benchmarks, for example Terminal Bench 2.1 (52.10 vs 43.82 / 26.52), NL2Repo-Bench (35.65 vs 28.77 / 16.55) and LongBench v2 (62.03 vs 48.12 / 47.99)."],"has_safetensors":true,"has_weights":true,"card_len":17392,"used_storage":38883686769,"created":"2026-10-23T08:47:29Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/GrootTeam/Groot-19B-A3B-261023","likes":7,"downloads":31,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.98,"lane":"fresh","vetted":false,"quant_sizes":{"Q4":{"gb":11.7,"file":"Q4_K_M","repo":"GrootTeam/Groot-19B-A3B-261023-GGUF","kind":"GGUF","community":false},"Q8":{"gb":20.8,"file":"Q8_0","repo":"mradermacher/Groot-19B-A3B-261023-GGUF","kind":"GGUF","community":false},"FP16":{"gb":38.9,"file":"official BF16 safetensors","repo":"GrootTeam/Groot-19B-A3B-261023","kind":"safetensors","community":false}},"summary":"Groot is a family of open-weight reasoning models for complex real-world tasks. Mixture-of-experts, 19B total, 3B active (128 experts, 8 per token). Context: 256K tokens (config). License: apache-2.0. Vendor-reported: \"Groot-19B-A3B-261023 beats both Qwen3.6-35B-A3B and Gemma-4-26B-A4B on 12 of the 16 benchmarks, for example Terminal Bench 2.1 (52.10 vs 43.82 / 26.52), NL2Repo-Bench (35.65 vs 28.77 / 16.55) and LongBench v2 (62.03 vs 48.12 / 47.99).\"","est":{"weights":{"Q4":11.7,"Q8":20.8,"FP16":38.9},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"GrootTeam/Groot-19B-A3B-261023-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/Groot-19B-A3B-261023-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"GrootTeam/Groot-19B-A3B-261023","fmt":"safetensors","community":false}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":13.4,"Q8":22.5,"FP16":40.6},"fit":{"nv8":null,"gpu12":null,"gpu16":"Q4","nv24":"Q8","nv32":"Q8","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q4","amd24":"Q8","strix":"FP16","gpu24":"Q8","gpu32":"Q8","mac16":null,"mac24":"Q4","mac36":"Q8","mac48":"Q8","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q8","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":null,"mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"SubMaroon/Kitchoon-V3-26B-A4B","name":"Kitchoon-V3-26B-A4B","org":"SubMaroon","date":"2026-10-20","license":"apache-2.0","params":25805933872,"params_src":"safetensors","active":4000000000.0,"ctx":262144,"kv_bytes_tok":245760,"dtypes":["BF16"],"pipeline":"","modality":"","summary_what":"The Kitchoon series is designed as an RP- and ERP-oriented model, focusing on character development and featuring a neutral writing style.","arch":{"class":"Gemma4ForConditionalGeneration","type":"MoE","experts":128,"experts_active":8,"active":4000000000.0,"active_src":"name"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":5753,"used_storage":51644178190,"created":"2026-10-20T21:47:53Z","first_seen":"2026-10-22T23:27:46Z","fresh":true,"url":"https://huggingface.co/SubMaroon/Kitchoon-V3-26B-A4B","likes":8,"downloads":86,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.96,"lane":"fresh","vetted":false,"quant_sizes":{"Q8":{"gb":26.9,"file":"Q8_0","repo":"SubMaroon/Kitchoon-V3-26B-A4B-GGUF","kind":"GGUF","community":false},"Q4":{"gb":16.8,"file":"Q4_K_M","repo":"mradermacher/Kitchoon-V3-26B-A4B-i1-GGUF","kind":"GGUF","community":false},"FP16":{"gb":51.6,"file":"official BF16 safetensors","repo":"SubMaroon/Kitchoon-V3-26B-A4B","kind":"safetensors","community":false}},"summary":"The Kitchoon series is designed as an RP- and ERP-oriented model, focusing on character development and featuring a neutral writing style. Mixture-of-experts, 26B total, 4B active (128 experts, 8 per token). Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":16.8,"Q8":26.9,"FP16":51.6},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/Kitchoon-V3-26B-A4B-i1-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"SubMaroon/Kitchoon-V3-26B-A4B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"SubMaroon/Kitchoon-V3-26B-A4B","fmt":"safetensors","community":false}},"kv_gb":2.0,"kv_src":"config","need":{"Q4":19.8,"Q8":29.9,"FP16":54.6},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q8","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":"Q8","mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":"real","mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"bayliner1980/Qwen3-0.6B-MoE-A0.5B","name":"Qwen3-0.6B-MoE-A0.5B","org":"bayliner1980","date":"2026-10-23","license":"apache-2.0","params":596967424,"params_src":"safetensors","active":500000000.0,"ctx":40960,"kv_bytes_tok":114688,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"Qwen3MoeForCausalLM","type":"MoE","experts":32,"experts_active":16,"active":500000000.0,"active_src":"name"},"claims":["At 0.6B, attention and the 156M-parameter embedding / LM head dominate the work per token, and top-16 activates half of the MLP."],"has_safetensors":true,"has_weights":true,"card_len":8461,"used_storage":1205711778,"created":"2026-10-23T06:24:43Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/bayliner1980/Qwen3-0.6B-MoE-A0.5B","likes":5,"downloads":33,"finetune":true,"gated":false,"quants":[{"id":"bayliner1980/Qwen3-0.6B-MoE-A0.5B-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.93,"lane":"fresh","vetted":false,"quant_sizes":{"Q4":{"gb":0.4,"file":"Q4_K_M","repo":"bayliner1980/Qwen3-0.6B-MoE-A0.5B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":1.2,"file":"BF16","repo":"bayliner1980/Qwen3-0.6B-MoE-A0.5B-GGUF","kind":"GGUF","community":false}},"summary":"Text generation model from bayliner1980 (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 597M total, 500M active (32 experts, 16 per token). Context: 40K tokens (config). License: apache-2.0. Vendor-reported: \"At 0.6B, attention and the 156M-parameter embedding / LM head dominate the work per token, and top-16 activates half of the MLP.\"","ollama":"https://ollama.com/library/qwen3","est":{"weights":{"Q4":0.4,"Q8":0.6,"FP16":1.2},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"bayliner1980/Qwen3-0.6B-MoE-A0.5B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.6B params"},"FP16":{"kind":"real","file":"BF16","repo":"bayliner1980/Qwen3-0.6B-MoE-A0.5B-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.9,"kv_src":"config","need":{"Q4":2.3,"Q8":2.5,"FP16":3.1},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"incoai/MiniMax-M3.1-Flash-DFlash2","name":"MiniMax-M3.1-Flash-DFlash2","org":"incoai","date":"2026-10-19","license":"apache-2.0","params":2123883776,"params_src":"safetensors","active":null,"ctx":1048576,"kv_bytes_tok":20480,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"DFlash 2 is a block-diffusion drafter for speculative decoding.","arch":{"class":"DFlash2DraftModel","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":4502,"used_storage":4248063377,"created":"2026-10-19T10:20:01Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/incoai/MiniMax-M3.1-Flash-DFlash2","likes":12,"downloads":107,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.89,"lane":"fresh","vetted":false,"quant_sizes":{"FP16":{"gb":4.2,"file":"official BF16 safetensors","repo":"incoai/MiniMax-M3.1-Flash-DFlash2","kind":"safetensors","community":false}},"summary":"DFlash 2 is a block-diffusion drafter for speculative decoding. Dense, 2.1B params. Context: 1M tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/minimax-m3.1-flash","est":{"weights":{"Q4":1.3,"Q8":2.3,"FP16":4.2},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 2.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 2.1B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"incoai/MiniMax-M3.1-Flash-DFlash2","fmt":"safetensors","community":false}},"kv_gb":0.2,"kv_src":"config","need":{"Q4":2.5,"Q8":3.5,"FP16":5.4},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"ReadyArt/Wicked-Stuff-26B-A4B-v2.0-Final","name":"Wicked-Stuff-26B-A4B-v2.0-Final","org":"ReadyArt","date":"2026-10-23","license":"apache-2.0","params":25805933872,"params_src":"safetensors","active":4000000000.0,"ctx":262144,"kv_bytes_tok":245760,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"Gemma4ForConditionalGeneration","type":"MoE","experts":128,"experts_active":8,"active":4000000000.0,"active_src":"name"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":11745,"used_storage":51644178190,"created":"2026-10-23T14:08:10Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/ReadyArt/Wicked-Stuff-26B-A4B-v2.0-Final","likes":2,"downloads":27,"finetune":true,"gated":false,"quants":[{"id":"ReadyArt/Wicked-Stuff-26B-A4B-v2.0-Final-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.87,"lane":"fresh","vetted":false,"quant_sizes":{"Q4":{"gb":16.8,"file":"Q4_K_M","repo":"ReadyArt/Wicked-Stuff-26B-A4B-v2.0-Final-GGUF","kind":"GGUF","community":false},"Q8":{"gb":26.9,"file":"Q8_0","repo":"ReadyArt/Wicked-Stuff-26B-A4B-v2.0-Final-GGUF","kind":"GGUF","community":false},"FP16":{"gb":51.6,"file":"official BF16 safetensors","repo":"ReadyArt/Wicked-Stuff-26B-A4B-v2.0-Final","kind":"safetensors","community":false}},"summary":"Text generation model from ReadyArt (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 26B total, 4B active (128 experts, 8 per token). Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":16.8,"Q8":26.9,"FP16":51.6},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"ReadyArt/Wicked-Stuff-26B-A4B-v2.0-Final-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"ReadyArt/Wicked-Stuff-26B-A4B-v2.0-Final-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"ReadyArt/Wicked-Stuff-26B-A4B-v2.0-Final","fmt":"safetensors","community":false}},"kv_gb":2.0,"kv_src":"config","need":{"Q4":19.8,"Q8":29.9,"FP16":54.6},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q8","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":"Q8","mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":"real","mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"berryber09/FastH3-OmniRef-8step-pruned-w4a8","name":"FastH3-OmniRef-8step-pruned-w4a8","org":"berryber09","date":"2026-10-23","license":"minimax-h3-community-license","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"text-to-video","modality":"video","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":1857,"used_storage":16555980952,"created":"2026-10-23T20:17:01Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/berryber09/FastH3-OmniRef-8step-pruned-w4a8","likes":4,"downloads":19,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.85,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"Text to video model from berryber09 (HF pipeline tag; the card has no one-line description). License: minimax-h3-community-license.","est":null,"license_class":"restricted"},{"type":"model","id":"GreenBitAI/Ternary-Bonsai-2-27B-gbx","name":"Ternary-Bonsai-2-27B-gbx","org":"GreenBitAI","date":"2026-10-20","license":"apache-2.0","params":27359638768,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["F32","U32","F16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"Bonsai 2 ships without a draft model; this head starts from the multi-token prediction head of Qwen3.8-27B, which Bonsai 2 is built from, and is distilled on text generated by Bonsai 2 itself -- prose, code, chat and coding-agent sessions.","arch":{"type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":9482,"used_storage":8847395082,"created":"2026-10-20T18:06:37Z","first_seen":"2026-10-22T23:27:46Z","fresh":true,"url":"https://huggingface.co/GreenBitAI/Ternary-Bonsai-2-27B-gbx","likes":2,"downloads":78,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.79,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"Bonsai 2 ships without a draft model; this head starts from the multi-token prediction head of Qwen3.8-27B, which Bonsai 2 is built from, and is distilled on text generated by Bonsai 2 itself -- prose, code, chat and coding-agent sessions. Dense, 27B params. Context: 256K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/bonsai-2","est":{"weights":{"Q4":16.4,"Q8":29.1,"FP16":54.7},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 27.4B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 27.4B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 27.4B params"}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":19.5,"Q8":32.2,"FP16":57.8},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":null,"amd24":"estimate","strix":"estimate","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"Phase-Technologies/Densy-330M-Chat","name":"Densy-330M-Chat","org":"Phase-Technologies","date":"2026-10-23","license":"apache-2.0","params":487073280,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"Densy-330M-Chat is an open-source, pure dense causal transformer language model designed and trained from scratch by Phase Technologies.","arch":{"class":"DensyForCausalLM"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":8994,"used_storage":1631247511,"created":"2026-10-23T16:01:47Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/Phase-Technologies/Densy-330M-Chat","likes":2,"downloads":22,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.77,"lane":"fresh","vetted":false,"quant_sizes":{"FP16":{"gb":1.0,"file":"official BF16 safetensors","repo":"Phase-Technologies/Densy-330M-Chat","kind":"safetensors","community":false}},"summary":"Densy-330M-Chat is an open-source, pure dense causal transformer language model designed and trained from scratch by Phase Technologies. 487M params. License: apache-2.0.","est":{"weights":{"Q4":0.3,"Q8":0.5,"FP16":1.0},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.5B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.5B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"Phase-Technologies/Densy-330M-Chat","fmt":"safetensors","community":false}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.8,"Q8":2.0,"FP16":2.5},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"owensong/Inflect-Mini-v3-Preview","name":"Inflect-Mini-v3-Preview","org":"owensong","date":"2026-10-23","license":"","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"text-to-speech","modality":"audio","summary_what":"","arch":{},"claims":[],"has_safetensors":false,"has_weights":true,"card_len":19570,"used_storage":148940381,"created":"2026-10-23T03:17:17Z","first_seen":"2026-10-12T20:10:36Z","fresh":true,"url":"https://huggingface.co/owensong/Inflect-Mini-v3-Preview","likes":5,"downloads":28,"finetune":false,"gated":true,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.75,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"Text to speech model from owensong (HF pipeline tag; the card has no one-line description).","est":null,"license_class":"unknown"},{"type":"model","id":"tencent/Hy-Nomas","name":"Hy-Nomas","org":"tencent","date":"2026-10-21","license":"apache-2.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"Hy-Nomas is a video world generation model that supports free-form, multi-turn interaction through text instructions and camera control.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19711,"used_storage":196404270231,"created":"2026-10-21T02:17:37Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/tencent/Hy-Nomas","likes":55,"downloads":0,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.59,"lane":"fresh","vetted":true,"quant_sizes":{},"summary":"Hy-Nomas is a video world generation model that supports free-form, multi-turn interaction through text instructions and camera control. License: apache-2.0.","est":null,"license_class":"permissive"},{"type":"model","id":"DavidAU/Qwen3.8-31B-Looping-Transformer","name":"Qwen3.8-31B-Looping-Transformer","org":"DavidAU","date":"2026-10-23","license":"apache-2.0","params":30825577520,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":4172,"used_storage":61664128278,"created":"2026-10-23T06:34:07Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/DavidAU/Qwen3.8-31B-Looping-Transformer","likes":17,"downloads":2,"finetune":true,"gated":true,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.47,"lane":"fresh","vetted":false,"quant_sizes":{"FP16":{"gb":61.7,"file":"official BF16 safetensors","repo":"DavidAU/Qwen3.8-31B-Looping-Transformer","kind":"safetensors","community":false}},"summary":"Image text to text model from DavidAU (HF pipeline tag; the card has no one-line description). 31B params. License: apache-2.0.","ollama":"https://ollama.com/library/qwen3.8","est":{"weights":{"Q4":18.5,"Q8":32.8,"FP16":61.7},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 30.8B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 30.8B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"DavidAU/Qwen3.8-31B-Looping-Transformer","fmt":"safetensors","community":false}},"kv_gb":2.6,"kv_src":"rough","need":{"Q4":22.1,"Q8":36.4,"FP16":65.3},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q4","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"estimate","strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"estimate","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"ReadyArt/Wicked-Stuff-31B-Final","name":"Wicked-Stuff-31B-Final","org":"ReadyArt","date":"2026-10-23","license":"apache-2.0","params":31273086512,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":983040,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"Gemma4ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":11571,"used_storage":62578507890,"created":"2026-10-23T06:54:38Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/ReadyArt/Wicked-Stuff-31B-Final","likes":2,"downloads":17,"finetune":true,"gated":false,"quants":[{"id":"ReadyArt/Wicked-Stuff-31B-Final-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.47,"lane":"fresh","vetted":false,"quant_sizes":{"Q4":{"gb":19.4,"file":"Q4_K_S","repo":"ReadyArt/Wicked-Stuff-31B-Final-GGUF","kind":"GGUF","community":false},"Q8":{"gb":34.0,"file":"Q8_0","repo":"ReadyArt/Wicked-Stuff-31B-Final-GGUF","kind":"GGUF","community":false},"FP16":{"gb":62.5,"file":"official BF16 safetensors","repo":"ReadyArt/Wicked-Stuff-31B-Final","kind":"safetensors","community":false}},"summary":"Text generation model from ReadyArt (HF pipeline tag; the card has no one-line description). Dense, 31B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":19.4,"Q8":34.0,"FP16":62.5},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_S","repo":"ReadyArt/Wicked-Stuff-31B-Final-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"ReadyArt/Wicked-Stuff-31B-Final-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"ReadyArt/Wicked-Stuff-31B-Final","fmt":"safetensors","community":false}},"kv_gb":8.1,"kv_src":"config","need":{"Q4":28.5,"Q8":43.1,"FP16":71.6},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":null,"strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":"Q4","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":null,"cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":null,"cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"mingqiancha233/Socrates-Image-1.0-31B","name":"Socrates-Image-1.0-31B","org":"mingqiancha233","date":"2026-10-22","license":"qwen-research","params":31000000000.0,"params_src":"name","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"text-to-image","modality":"image","summary_what":"Socrates-Image-1.0 is a self-contained dense text-to-image model with recurrent visual-semantic feedback.","arch":{"class":"SocratesImageModel"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":5543,"used_storage":62973612078,"created":"2026-10-22T19:41:09Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/mingqiancha233/Socrates-Image-1.0-31B","likes":2,"downloads":22,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.46,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"Socrates-Image-1.0 is a self-contained dense text-to-image model with recurrent visual-semantic feedback. 31B params. License: qwen-research.","est":{"weights":{"Q4":18.6,"Q8":32.9,"FP16":62.0},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 31.0B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 31.0B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 31.0B params"}},"kv_gb":2.6,"kv_src":"rough","need":{"Q4":22.2,"Q8":36.5,"FP16":65.6},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q4","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":null,"amd24":"estimate","strix":"estimate","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"tencent/Hy-Nomas-Diffusers","name":"Hy-Nomas-Diffusers","org":"tencent","date":"2026-10-24","license":"apache-2.0","params":8553434592,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"","modality":"","summary_what":"Hy-Nomas is a video world generation model that supports free-form, multi-turn interaction through text instructions and camera control.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":15721,"used_storage":45203967145,"created":"2026-10-24T05:45:20Z","first_seen":"2026-10-24T11:24:09Z","fresh":true,"url":"https://huggingface.co/tencent/Hy-Nomas-Diffusers","likes":6,"downloads":2,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.45,"lane":"fresh","vetted":true,"quant_sizes":{},"summary":"Hy-Nomas is a video world generation model that supports free-form, multi-turn interaction through text instructions and camera control. 8.6B params. License: apache-2.0.","est":{"weights":{"Q4":5.1,"Q8":9.1,"FP16":17.1},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 8.6B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 8.6B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 8.6B params"}},"kv_gb":0.7,"kv_src":"rough","need":{"Q4":6.8,"Q8":10.8,"FP16":18.8},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"Q8","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"BlueRoseLabs/Silencio","name":"Silencio","org":"BlueRoseLabs","date":"2026-10-23","license":"apache-2.0","params":33607366,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"text-to-speech","modality":"audio","summary_what":"Silencio is built on two open models.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":13458,"used_storage":177661332,"created":"2026-10-23T16:11:23Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/BlueRoseLabs/Silencio","likes":2,"downloads":12,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.45,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"Silencio is built on two open models. 34M params. License: apache-2.0.","est":{"weights":{"Q4":0.0,"Q8":0.0,"FP16":0.1},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.0B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.0B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.0B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.5,"Q8":1.5,"FP16":1.6},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"ReadyArt/Omega-Arsenal-26B-A4B-Final","name":"Omega-Arsenal-26B-A4B-Final","org":"ReadyArt","date":"2026-10-22","license":"apache-2.0","params":25805936206,"params_src":"safetensors","active":4000000000.0,"ctx":262144,"kv_bytes_tok":245760,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"Gemma4ForConditionalGeneration","type":"MoE","experts":128,"experts_active":8,"active":4000000000.0,"active_src":"name"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":11774,"used_storage":51644179542,"created":"2026-10-22T21:35:03Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/ReadyArt/Omega-Arsenal-26B-A4B-Final","likes":2,"downloads":16,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.36,"lane":"fresh","vetted":false,"quant_sizes":{"Q4":{"gb":17.7,"file":"Q4_K_M","repo":"ReadyArt/Omega-Arsenal-26B-A4B-Final-GGUF","kind":"GGUF","community":false},"Q8":{"gb":27.6,"file":"Q8_0","repo":"ReadyArt/Omega-Arsenal-26B-A4B-Final-GGUF","kind":"GGUF","community":false},"FP16":{"gb":51.6,"file":"official BF16 safetensors","repo":"ReadyArt/Omega-Arsenal-26B-A4B-Final","kind":"safetensors","community":false}},"summary":"Text generation model from ReadyArt (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 26B total, 4B active (128 experts, 8 per token). Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":17.7,"Q8":27.6,"FP16":51.6},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"ReadyArt/Omega-Arsenal-26B-A4B-Final-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"ReadyArt/Omega-Arsenal-26B-A4B-Final-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"ReadyArt/Omega-Arsenal-26B-A4B-Final","fmt":"safetensors","community":false}},"kv_gb":2.0,"kv_src":"config","need":{"Q4":20.7,"Q8":30.6,"FP16":54.6},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q8","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":"Q8","mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":"real","mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"h2oai/timeh2o-v1","name":"timeh2o-v1","org":"h2oai","date":"2026-10-22","license":"h2o-model-non-commercial-licence-v1.0","params":108764558,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"time-series-forecasting","modality":"other","summary_what":"Time series foundation model by H2O.ai.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":1411,"used_storage":435111800,"created":"2026-10-22T15:00:56Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/h2oai/timeh2o-v1","likes":7,"downloads":13,"finetune":false,"gated":true,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.35,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"Time series foundation model by H2O.ai. 109M params. License: h2o-model-non-commercial-licence-v1.0.","est":{"weights":{"Q4":0.1,"Q8":0.1,"FP16":0.2},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.1B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.1B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.6,"Q8":1.6,"FP16":1.7},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"DavidAU/Qwen3.8-40B-Ultimate","name":"Qwen3.8-40B-Ultimate","org":"DavidAU","date":"2026-10-22","license":"apache-2.0","params":39958026224,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":3961,"used_storage":79929068704,"created":"2026-10-22T00:07:26Z","first_seen":"2026-10-22T11:23:11Z","fresh":true,"url":"https://huggingface.co/DavidAU/Qwen3.8-40B-Ultimate","likes":14,"downloads":1,"finetune":true,"gated":true,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.21,"lane":"fresh","vetted":false,"quant_sizes":{"FP16":{"gb":79.9,"file":"official BF16 safetensors","repo":"DavidAU/Qwen3.8-40B-Ultimate","kind":"safetensors","community":false}},"summary":"Image text to text model from DavidAU (HF pipeline tag; the card has no one-line description). 40B params. License: apache-2.0.","ollama":"https://ollama.com/library/qwen3.8","est":{"weights":{"Q4":24.0,"Q8":42.5,"FP16":79.9},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 40.0B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 40.0B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"DavidAU/Qwen3.8-40B-Ultimate","fmt":"safetensors","community":false}},"kv_gb":3.4,"kv_src":"rough","need":{"Q4":28.4,"Q8":46.9,"FP16":84.3},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":null,"strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":"Q4","mac64":"Q8","mac96":"Q8","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":null,"cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"estimate","nv48":"estimate","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":null,"cpu64":"estimate","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"ATH-MaaS/PixText","name":"PixText","org":"ATH-MaaS","date":"2026-10-22","license":"apache-2.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"Use models/PixText-Teacher to load Teacher.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":12806,"used_storage":3859967609,"created":"2026-10-22T06:41:17Z","first_seen":"2026-10-23T23:18:57Z","fresh":true,"url":"https://huggingface.co/ATH-MaaS/PixText","likes":13,"downloads":0,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.2,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"Use models/PixText-Teacher to load Teacher. License: apache-2.0.","est":null,"license_class":"permissive"},{"type":"model","id":"ReadyArt/Wicked-Stuff-26B-A4B-v2.1-FINAL","name":"Wicked-Stuff-26B-A4B-v2.1-FINAL","org":"ReadyArt","date":"2026-10-24","license":"apache-2.0","params":25805933872,"params_src":"safetensors","active":4000000000.0,"ctx":262144,"kv_bytes_tok":245760,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"Gemma4ForConditionalGeneration","type":"MoE","experts":128,"experts_active":8,"active":4000000000.0,"active_src":"name"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":11669,"used_storage":51644178190,"created":"2026-10-24T12:19:53Z","first_seen":"2026-10-24T23:23:29Z","fresh":true,"url":"https://huggingface.co/ReadyArt/Wicked-Stuff-26B-A4B-v2.1-FINAL","likes":2,"downloads":0,"finetune":true,"gated":false,"quants":[{"id":"ReadyArt/Wicked-Stuff-26B-A4B-v2.1-Final-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.18,"lane":"fresh","vetted":false,"quant_sizes":{"Q4":{"gb":15.5,"file":"Q4_K_S","repo":"ReadyArt/Wicked-Stuff-26B-A4B-v2.1-Final-GGUF","kind":"GGUF","community":false},"Q8":{"gb":26.9,"file":"Q8_0","repo":"ReadyArt/Wicked-Stuff-26B-A4B-v2.1-Final-GGUF","kind":"GGUF","community":false},"FP16":{"gb":51.6,"file":"official BF16 safetensors","repo":"ReadyArt/Wicked-Stuff-26B-A4B-v2.1-FINAL","kind":"safetensors","community":false}},"summary":"Text generation model from ReadyArt (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 26B total, 4B active (128 experts, 8 per token). Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":15.5,"Q8":26.9,"FP16":51.6},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_S","repo":"ReadyArt/Wicked-Stuff-26B-A4B-v2.1-Final-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"ReadyArt/Wicked-Stuff-26B-A4B-v2.1-Final-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"ReadyArt/Wicked-Stuff-26B-A4B-v2.1-FINAL","fmt":"safetensors","community":false}},"kv_gb":2.0,"kv_src":"config","need":{"Q4":18.5,"Q8":29.9,"FP16":54.6},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q8","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":"Q8","mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":"real","mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"hi3d-official/Hi3Dv1.0","name":"Hi3Dv1.0","org":"hi3d-official","date":"2026-10-16","license":"hi3d-open-license-agreement-1.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"image-to-3d","modality":"other","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":15400,"used_storage":7735349555,"created":"2026-10-16T07:51:15Z","first_seen":"2026-10-23T11:26:00Z","fresh":true,"url":"https://huggingface.co/hi3d-official/Hi3Dv1.0","likes":38,"downloads":0,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.18,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"Image to 3d model from hi3d-official (HF pipeline tag; the card has no one-line description). License: hi3d-open-license-agreement-1.0.","est":null,"license_class":"restricted"},{"type":"model","id":"e-n-v-y/None-qwenimage2.1-vae-comfyui","name":"None-qwenimage2.1-vae-comfyui","org":"e-n-v-y","date":"2026-10-21","license":"qwen-research","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19831,"used_storage":1362461138,"created":"2026-10-21T00:39:14Z","first_seen":"2026-10-22T11:23:11Z","fresh":true,"url":"https://huggingface.co/e-n-v-y/None-qwenimage2.1-vae-comfyui","likes":14,"downloads":0,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.15,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"License: qwen-research.","ollama":"https://ollama.com/library/qwen","est":null,"license_class":"restricted"},{"type":"model","id":"nikopueringer/CorridorKey2.0","name":"CorridorKey2.0","org":"nikopueringer","date":"2026-10-13","license":"corridor-key-licence-2.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"image-segmentation","modality":"other","summary_what":"CorridorKey runs on features from three frozen foundation models.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":3119,"used_storage":2661717584,"created":"2026-10-13T00:37:33Z","first_seen":"2026-10-22T11:23:11Z","fresh":true,"url":"https://huggingface.co/nikopueringer/CorridorKey2.0","likes":25,"downloads":0,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.09,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"CorridorKey runs on features from three frozen foundation models. License: corridor-key-licence-2.0.","est":null,"license_class":"restricted"},{"type":"model","id":"AnveshAI/AnveshAI-CoderLM-0.1B","name":"AnveshAI-CoderLM-0.1B","org":"AnveshAI","date":"2026-10-22","license":"apache-2.0","params":100176321,"params_src":"safetensors","active":null,"ctx":2048,"kv_bytes_tok":7680,"dtypes":["F16"],"pipeline":"text-generation","modality":"text","summary_what":"CoderLM is a compact 100.2M-parameter language model purpose-built for multi-language code generation across 13 programming and markup languages \u2014 Python, Java, JavaScript, C++, C, Rust, TypeScript, Go, HTML, CSS, Shell, SQL, and Markdown.","arch":{"class":"AnveshAICoderLMForCausalLM","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":12903,"used_storage":201005526,"created":"2026-10-22T18:45:22Z","first_seen":"2026-10-24T23:23:29Z","fresh":true,"url":"https://huggingface.co/AnveshAI/AnveshAI-CoderLM-0.1B","likes":3,"downloads":0,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.06,"lane":"fresh","vetted":false,"quant_sizes":{"FP16":{"gb":0.2,"file":"official F16 safetensors","repo":"AnveshAI/AnveshAI-CoderLM-0.1B","kind":"safetensors","community":false}},"summary":"CoderLM is a compact 100.2M-parameter language model purpose-built for multi-language code generation across 13 programming and markup languages \u2014 Python, Java, JavaScript, C++, C, Rust, TypeScript, Go, HTML, CSS, Shell, SQL, and Markdown. Dense, 100M params. Context: 2K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":0.1,"Q8":0.1,"FP16":0.2},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.1B params"},"FP16":{"kind":"real","file":"official F16 safetensors","repo":"AnveshAI/AnveshAI-CoderLM-0.1B","fmt":"safetensors","community":false}},"kv_gb":0.1,"kv_src":"config","need":{"Q4":1.2,"Q8":1.2,"FP16":1.3},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"turion-ai/turion-jev-31b","name":"turion-jev-31b","org":"turion-ai","date":"2026-10-22","license":"apache-2.0","params":31273086512,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":983040,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"Turion Jev 31B is a decision model by Turion AI, fine-tuned from google/gemma-4-31B-it on the Turion AI model training platform.","arch":{"class":"Gemma4ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":5940,"used_storage":62578506074,"created":"2026-10-22T16:31:59Z","first_seen":"2026-10-24T11:24:09Z","fresh":true,"url":"https://huggingface.co/turion-ai/turion-jev-31b","likes":2,"downloads":0,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.04,"lane":"fresh","vetted":false,"quant_sizes":{"FP16":{"gb":62.5,"file":"official BF16 safetensors","repo":"turion-ai/turion-jev-31b","kind":"safetensors","community":false}},"summary":"Turion Jev 31B is a decision model by Turion AI, fine-tuned from google/gemma-4-31B-it on the Turion AI model training platform. Dense, 31B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":18.8,"Q8":33.2,"FP16":62.5},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 31.3B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 31.3B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"turion-ai/turion-jev-31b","fmt":"safetensors","community":false}},"kv_gb":8.1,"kv_src":"config","need":{"Q4":27.9,"Q8":42.3,"FP16":71.6},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":null,"strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":"Q4","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":null,"cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"estimate","nv48":"estimate","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":"estimate","mac64":"estimate","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":null,"cpu64":"estimate","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"Luffuly/streaming-code-world-model","name":"streaming-code-world-model","org":"Luffuly","date":"2026-10-11","license":"","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":1059,"used_storage":2385650968,"created":"2026-10-11T20:00:44Z","first_seen":"2026-10-24T11:24:09Z","fresh":true,"url":"https://huggingface.co/Luffuly/streaming-code-world-model","likes":11,"downloads":0,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"","velocity":0.03,"lane":"fresh","vetted":false,"quant_sizes":{},"summary":"","est":null,"license_class":"unknown"},{"type":"model","id":"RecursiveSI/flint","name":"flint","org":"RecursiveSI","date":"2026-10-21","license":"apache-2.0","params":39785287456,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":100352,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"Flint is Recursive\u2019s mixture-of-experts language model for coding agents, with 39.8 billion total parameters and approximately 2.5 billion active per token.","arch":{"class":"FlintForCausalLM","type":"MoE","experts":256,"experts_active":6},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":6147,"used_storage":79587116933,"created":"2026-10-21T14:20:58Z","first_seen":"2026-10-21T23:23:07Z","fresh":false,"url":"https://huggingface.co/RecursiveSI/flint","likes":770,"downloads":4795,"finetune":false,"gated":false,"quants":[{"id":"bartowski/RecursiveSI_flint-GGUF","kind":"GGUF"},{"id":"ggml-org/Flint-GGUF","kind":"GGUF"},{"id":"lmstudio-community/flint-GGUF","kind":"GGUF"},{"id":"mlx-community/flint-OptiQ-4bit","kind":"MLX"},{"id":"unsloth/flint-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"770 likes","velocity":68.67,"lane":"main","quant_sizes":{"Q4":{"gb":25.1,"file":"UD-Q4_K_XL","repo":"unsloth/flint-GGUF","kind":"GGUF","community":false},"Q8":{"gb":42.3,"file":"Q8_0","repo":"unsloth/flint-GGUF","kind":"GGUF","community":false},"FP16":{"gb":79.6,"file":"BF16","repo":"unsloth/flint-GGUF","kind":"GGUF","community":false}},"summary":"Flint is Recursive\u2019s mixture-of-experts language model for coding agents, with 39.8 billion total parameters and approximately 2.5 billion active per token. Mixture-of-experts, 40B total (256 experts, 6 per token). Context: 256K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/flint","est":{"weights":{"Q4":25.1,"Q8":42.3,"FP16":79.6},"weights_src":{"Q4":{"kind":"real","file":"UD-Q4_K_XL","repo":"unsloth/flint-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"unsloth/flint-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"unsloth/flint-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.8,"kv_src":"config","need":{"Q4":26.9,"Q8":44.1,"FP16":81.4},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":null,"strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q4","mac64":"Q8","mac96":"Q8","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":null,"cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":null,"cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"KRAFTON/Raon-1-mini-preview","name":"Raon-1-mini-preview","org":"KRAFTON","date":"2026-10-21","license":"apache-2.0","params":28783218402,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":28800,"dtypes":["BF16","F32","F8_E4M3"],"pipeline":"text-generation","modality":"text","summary_what":"Raon-1 mini is a 30B-A4B Mixture-of-Experts (MoE) language model, designed and trained from scratch by KRAFTON AI.","arch":{"class":"Raon1MiniForCausalLM","type":"MoE","experts":128,"experts_active":6},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19355,"used_storage":32714641804,"created":"2026-10-21T00:14:48Z","first_seen":"2026-10-21T11:18:13Z","fresh":false,"url":"https://huggingface.co/KRAFTON/Raon-1-mini-preview","likes":163,"downloads":917,"finetune":false,"gated":false,"quants":[{"id":"KRAFTON/Raon-1-mini-preview-gguf","kind":"GGUF"},{"id":"KRAFTON/Raon-1-mini-preview-mlx","kind":"MLX"},{"id":"KRAFTON/Raon-1-mini-preview-nvfp4","kind":"quant"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"163 likes","velocity":11.35,"lane":"main","quant_sizes":{"Q4":{"gb":19.7,"file":"Q4_K_M","repo":"KRAFTON/Raon-1-mini-preview-gguf","kind":"GGUF","community":false},"Q8":{"gb":31.4,"file":"Q8_0","repo":"KRAFTON/Raon-1-mini-preview-gguf","kind":"GGUF","community":false}},"summary":"Raon-1 mini is a 30B-A4B Mixture-of-Experts (MoE) language model, designed and trained from scratch by KRAFTON AI. Mixture-of-experts, 29B total (128 experts, 6 per token). Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":19.7,"Q8":31.4,"FP16":57.6},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"KRAFTON/Raon-1-mini-preview-gguf","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"KRAFTON/Raon-1-mini-preview-gguf","fmt":"GGUF","community":false},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 28.8B params"}},"kv_gb":0.2,"kv_src":"config","need":{"Q4":20.9,"Q8":32.6,"FP16":58.8},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":null,"amd24":"real","strix":"estimate","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"real","cpu64":"real","cpu128":"estimate","cpu256":"estimate"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"inclusionAI/Ling-3.0-tiny-UBTree","name":"Ling-3.0-tiny-UBTree","org":"inclusionAI","date":"2026-10-21","license":"mit","params":183127326,"params_src":"safetensors","active":null,"ctx":131072,"kv_bytes_tok":40960,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"Ling3UBTreeModel","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":5051,"used_storage":366492459,"created":"2026-10-21T13:49:29Z","first_seen":"2026-10-21T23:23:07Z","fresh":false,"url":"https://huggingface.co/inclusionAI/Ling-3.0-tiny-UBTree","likes":12,"downloads":241,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":3.1,"lane":"main","quant_sizes":{"FP16":{"gb":0.4,"file":"official BF16 safetensors","repo":"inclusionAI/Ling-3.0-tiny-UBTree","kind":"safetensors","community":false}},"summary":"Text generation model from inclusionAI (HF pipeline tag; the card has no one-line description). Dense, 183M params. Context: 128K tokens (config). License: mit.","est":{"weights":{"Q4":0.1,"Q8":0.2,"FP16":0.4},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.2B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"inclusionAI/Ling-3.0-tiny-UBTree","fmt":"safetensors","community":false}},"kv_gb":0.3,"kv_src":"config","need":{"Q4":1.4,"Q8":1.5,"FP16":1.7},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"Comfy-Org/Fibo-2.0","name":"Fibo-2.0","org":"Comfy-Org","date":"2026-10-20","license":"cc-by-nc-4.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":2143,"used_storage":65782660712,"created":"2026-10-20T18:53:36Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/Comfy-Org/Fibo-2.0","likes":161,"downloads":111611,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"161 likes, 111611 downloads","velocity":1112.18,"lane":"main","quant_sizes":{},"summary":"License: cc-by-nc-4.0.","est":null,"license_class":"restricted"},{"type":"model","id":"FastVideo/FastH3-Omni-Comfy","name":"FastH3-Omni-Comfy","org":"FastVideo","date":"2026-10-20","license":"minimax-h3-community-license-agreement","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"FastH3 Omni was built in close collaboration with Nuva Lab, which co-developed the model with FastVideo, defined and curated the omni reference dataset, and brought production-grounded creative video workflows to FastH3.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":4688,"used_storage":169413412215,"created":"2026-10-20T04:58:26Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/FastVideo/FastH3-Omni-Comfy","likes":131,"downloads":154490,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"131 likes, 154490 downloads","velocity":1351.37,"lane":"main","quant_sizes":{"Q4":{"gb":12.7,"file":"Q4_K_M","repo":"Abiray/FastH3-Omni-GGUF","kind":"GGUF","community":true}},"summary":"FastH3 Omni was built in close collaboration with Nuva Lab, which co-developed the model with FastVideo, defined and curated the omni reference dataset, and brought production-grounded creative video workflows to FastH3. License: minimax-h3-community-license-agreement.","est":{"weights":{"Q4":12.7},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"Abiray/FastH3-Omni-GGUF","fmt":"GGUF","community":true}},"kv_gb":1.0,"kv_src":"rough","need":{"Q4":14.7},"fit":{"nv8":null,"gpu12":null,"gpu16":"Q4","nv24":"Q4","nv32":"Q4","nv48":"Q4","nv96":"Q4","spark1":"Q4","spark2":"Q4","amd16":"Q4","amd24":"Q4","strix":"Q4","gpu24":null,"gpu32":null,"mac16":null,"mac24":"Q4","mac36":"Q4","mac48":"Q4","mac64":"Q4","mac96":"Q4","mac128":"Q4","mac192":"Q4","mac256":"Q4","mac512":"Q4","cpu32":"Q4","cpu64":"Q4","cpu128":"Q4","cpu256":"Q4"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"restricted"},{"type":"model","id":"XHToken/Spark-X2.6-4B","name":"Spark-X2.6-4B","org":"XHToken","date":"2026-10-19","license":"apache-2.0","params":4112079360,"params_src":"safetensors","active":null,"ctx":1048576,"kv_bytes_tok":147456,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"Search for \"Spark-X2.6\" in LM Studio and download the model.","arch":{"class":"Spark2_5ForCausalLM","type":"dense"},"claims":["They achieve state-of-the-art performance among similarly sized open-source models, matching or even surpassing models with approximately 9B parameters."],"has_safetensors":true,"has_weights":true,"card_len":13932,"used_storage":12686523518,"created":"2026-10-19T03:19:51Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/XHToken/Spark-X2.6-4B","likes":443,"downloads":2537,"finetune":true,"gated":false,"quants":[{"id":"XHToken/Spark-X2.6-4B-FP8","kind":"quant"},{"id":"XHToken/Spark-X2.6-4B-GGUF","kind":"GGUF"},{"id":"XHToken/Spark-X2.6-4B-INT4","kind":"quant"},{"id":"XHToken/Spark-X2.6-4B-INT8","kind":"quant"},{"id":"bartowski/Spark-X2.6-4B-GGUF","kind":"GGUF"},{"id":"mlx-community/Spark-X2.6-4B-OptiQ-4bit","kind":"MLX"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"443 likes","velocity":21.28,"lane":"main","quant_sizes":{"Q4":{"gb":3.5,"file":"Q4_K_M","repo":"XHToken/Spark-X2.6-4B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":4.4,"file":"Q8_0","repo":"XHToken/Spark-X2.6-4B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":8.2,"file":"BF16","repo":"bartowski/Spark-X2.6-4B-GGUF","kind":"GGUF","community":false}},"summary":"Search for \"Spark-X2.6\" in LM Studio and download the model. Dense, 4.1B params. Context: 1M tokens (config). License: apache-2.0. Vendor-reported: \"They achieve state-of-the-art performance among similarly sized open-source models, matching or even surpassing models with approximately 9B parameters.\"","est":{"weights":{"Q4":3.5,"Q8":4.4,"FP16":8.2},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"XHToken/Spark-X2.6-4B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"XHToken/Spark-X2.6-4B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"bartowski/Spark-X2.6-4B-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.2,"kv_src":"config","need":{"Q4":5.7,"Q8":6.6,"FP16":10.4},"fit":{"nv8":"Q8","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"lodestones/Kromoss","name":"Kromoss","org":"lodestones","date":"2026-10-19","license":"apache-2.0","params":5082308624,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16","F32"],"pipeline":"text-to-image","modality":"image","summary_what":"Kromoss is a full fine-tune of MOSS-Image-Base at 1024 px.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":7493,"used_storage":188035853247,"created":"2026-10-19T01:34:35Z","first_seen":"2026-10-19T11:24:28Z","fresh":false,"url":"https://huggingface.co/lodestones/Kromoss","likes":246,"downloads":21143,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"246 likes, 21143 downloads","velocity":150.82,"lane":"main","quant_sizes":{"Q4":{"gb":3.6,"file":"Q4_K_M","repo":"molbal/Kromoss-GGUF","kind":"GGUF","community":true},"Q8":{"gb":5.7,"file":"Q8_0","repo":"molbal/Kromoss-GGUF","kind":"GGUF","community":true}},"summary":"Kromoss is a full fine-tune of MOSS-Image-Base at 1024 px. 5.1B params. License: apache-2.0.","est":{"weights":{"Q4":3.6,"Q8":5.7,"FP16":10.2},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"molbal/Kromoss-GGUF","fmt":"GGUF","community":true},"Q8":{"kind":"real","file":"Q8_0","repo":"molbal/Kromoss-GGUF","fmt":"GGUF","community":true},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 5.1B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":5.1,"Q8":7.2,"FP16":11.7},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"briaai/Fibo-2.0-edit-turbo","name":"Fibo-2.0-edit-turbo","org":"briaai","date":"2026-10-19","license":"bria-fibo","params":8127342213,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"image-to-image","modality":"image","summary_what":"Fibo 2.0 is the of Bria's FIBO family.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":13731,"used_storage":25485587073,"created":"2026-10-19T13:11:51Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/briaai/Fibo-2.0-edit-turbo","likes":111,"downloads":886,"finetune":true,"gated":true,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"111 likes","velocity":7.66,"lane":"main","quant_sizes":{},"summary":"Fibo 2.0 is the of Bria's FIBO family. 8.1B params. License: bria-fibo.","est":{"weights":{"Q4":4.9,"Q8":8.6,"FP16":16.3},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 8.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 8.1B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 8.1B params"}},"kv_gb":0.7,"kv_src":"rough","need":{"Q4":6.6,"Q8":10.3,"FP16":18.0},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"gradium/phonon-31m-en","name":"phonon-31m-en","org":"gradium","date":"2026-10-19","license":"apache-2.0","params":41096035,"params_src":"safetensors","active":null,"ctx":8192,"kv_bytes_tok":12288,"dtypes":["F32","BF16"],"pipeline":"text-to-speech","modality":"audio","summary_what":"phonon-31m is a text-to-speech model from the Phonon family.","arch":{"class":"GradiumPhononTTSForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":10288,"used_storage":222988757,"created":"2026-10-19T12:02:53Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/gradium/phonon-31m-en","likes":100,"downloads":20631,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"100 likes, 20631 downloads","velocity":157.84,"lane":"main","quant_sizes":{"FP16":{"gb":0.1,"file":"BF16","repo":"danyalmalik/phonon-31m-en-GGUF","kind":"GGUF","community":true}},"summary":"phonon-31m is a text-to-speech model from the Phonon family. Dense, 41M params. Context: 8K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/phonon","est":{"weights":{"Q4":0.0,"Q8":0.0,"FP16":0.1},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.0B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.0B params"},"FP16":{"kind":"real","file":"BF16","repo":"danyalmalik/phonon-31m-en-GGUF","fmt":"GGUF","community":true}},"kv_gb":0.1,"kv_src":"config","need":{"Q4":1.1,"Q8":1.1,"FP16":1.2},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"vllm-sr/d3-turbo","name":"d3-turbo","org":"vllm-sr","date":"2026-10-19","license":"apache-2.0","params":26085330160,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16"],"pipeline":"zero-shot-classification","modality":"other","summary_what":"d3-turbo is the upgraded 27B multimodal decision model of Decision 3.0, the decision models of vLLM Semantic Router.","arch":{"class":"Qwen3_5Model","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":13087,"used_storage":104371677886,"created":"2026-10-19T02:14:31Z","first_seen":"2026-10-19T23:17:53Z","fresh":false,"url":"https://huggingface.co/vllm-sr/d3-turbo","likes":88,"downloads":1456,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"88 likes","velocity":10.94,"lane":"main","quant_sizes":{"FP16":{"gb":52.2,"file":"official BF16 safetensors","repo":"vllm-sr/d3-turbo","kind":"safetensors","community":false}},"summary":"d3-turbo is the upgraded 27B multimodal decision model of Decision 3.0, the decision models of vLLM Semantic Router. Dense, 26B params. Context: 256K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/d3","est":{"weights":{"Q4":15.7,"Q8":27.7,"FP16":52.2},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 26.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 26.1B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"vllm-sr/d3-turbo","fmt":"safetensors","community":false}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":18.8,"Q8":30.8,"FP16":55.3},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q8","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":"Q8","mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"estimate","strix":"real","gpu24":null,"gpu32":"estimate","mac16":null,"mac24":null,"mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"gradium/phonon-40m-en","name":"phonon-40m-en","org":"gradium","date":"2026-10-19","license":"cc-by-nc-4.0","params":50539363,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32","BF16"],"pipeline":"text-to-speech","modality":"audio","summary_what":"phonon-40m is a text-to-speech model from the Phonon family.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":10343,"used_storage":257242077,"created":"2026-10-19T12:01:47Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/gradium/phonon-40m-en","likes":67,"downloads":2270,"finetune":false,"gated":true,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"67 likes","velocity":17.79,"lane":"main","quant_sizes":{"FP16":{"gb":0.1,"file":"official BF16 safetensors","repo":"gradium/phonon-40m-en","kind":"safetensors","community":false}},"summary":"phonon-40m is a text-to-speech model from the Phonon family. 51M params. License: cc-by-nc-4.0.","ollama":"https://ollama.com/library/phonon","est":{"weights":{"Q4":0.0,"Q8":0.1,"FP16":0.1},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.1B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"gradium/phonon-40m-en","fmt":"safetensors","community":false}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.5,"Q8":1.6,"FP16":1.6},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"nvidia/Cosmos3-H-Dreams","name":"Cosmos3-H-Dreams","org":"nvidia","date":"2026-10-19","license":"nvidia-open-model-license","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"image-to-video","modality":"video","summary_what":"Cosmos3-H-Dreams is the successor of Cosmos-H-Dreams (arXiv:2608.24199), which was based on Cosmos-Predict2.5.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":16883,"used_storage":60692765392,"created":"2026-10-19T12:35:12Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/nvidia/Cosmos3-H-Dreams","likes":8,"downloads":359,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":2.81,"lane":"main","quant_sizes":{},"summary":"Cosmos3-H-Dreams is the successor of Cosmos-H-Dreams (arXiv:2608.24199), which was based on Cosmos-Predict2.5. License: nvidia-open-model-license.","est":null,"license_class":"restricted"},{"type":"model","id":"LiquidAI/LFM2.5-24B-A2B-DSpark","name":"LFM2.5-24B-A2B-DSpark","org":"LiquidAI","date":"2026-10-19","license":"lfm1.0","params":327707521,"params_src":"safetensors","active":null,"ctx":131072,"kv_bytes_tok":10240,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"LFM2.5-24B-A2B-DSpark is a speculative-decoding draft model for LiquidAI/LFM2.5-24B-A2B.","arch":{"class":"Lfm2DSparkDraftModel","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":5503,"used_storage":655957832,"created":"2026-10-19T22:39:26Z","first_seen":"2026-10-21T11:18:13Z","fresh":false,"url":"https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-DSpark","likes":7,"downloads":660,"finetune":true,"gated":false,"quants":[{"id":"LiquidAI/LFM2.5-24B-A2B-DSpark-GGUF","kind":"GGUF"}],"x_mentions":[{"url":"https://x.com/jvr0x/status/2112817955138326575","user":"jvr0x","likes":2}],"x_likes":2,"x_posts":1,"signal":"trusted org","velocity":15.52,"lane":"main","quant_sizes":{"FP16":{"gb":0.7,"file":"F16","repo":"LiquidAI/LFM2.5-24B-A2B-DSpark-GGUF","kind":"GGUF","community":false}},"summary":"LFM2.5-24B-A2B-DSpark is a speculative-decoding draft model for LiquidAI/LFM2.5-24B-A2B. Dense, 328M params. Context: 128K tokens (config). License: lfm1.0.","ollama":"https://ollama.com/library/lfm2.5","est":{"weights":{"Q4":0.2,"Q8":0.3,"FP16":0.7},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.3B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.3B params"},"FP16":{"kind":"real","file":"F16","repo":"LiquidAI/LFM2.5-24B-A2B-DSpark-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.1,"kv_src":"config","need":{"Q4":1.3,"Q8":1.4,"FP16":1.8},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"nvidia/Cosmos3-H-Embodied-Nano","name":"Cosmos3-H-Embodied-Nano","org":"nvidia","date":"2026-10-19","license":"nvidia-open-model-license","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"robotics","modality":"robotics","summary_what":"Cosmos3-H-Embodied-Nano is the successor of Cosmos-H-Surgical-Simulator, which was a forward-dynamics-only surgical simulator built on the 2B-parameter Cosmos-Predict2.5 diffusion transformer.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19509,"used_storage":30346382696,"created":"2026-10-19T12:45:34Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/nvidia/Cosmos3-H-Embodied-Nano","likes":2,"downloads":96,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.75,"lane":"main","quant_sizes":{},"summary":"Cosmos3-H-Embodied-Nano is the successor of Cosmos-H-Surgical-Simulator, which was a forward-dynamics-only surgical simulator built on the 2B-parameter Cosmos-Predict2.5 diffusion transformer. License: nvidia-open-model-license.","est":null,"license_class":"restricted"},{"type":"model","id":"nvidia/cosmos3-nano-libero10-sdg-export-S2_s43","name":"cosmos3-nano-libero10-sdg-export-S2_s43","org":"nvidia","date":"2026-10-19","license":"nvidia-open-model-license","params":15173136576,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"robotics","modality":"robotics","summary_what":"","arch":{"class":"Cosmos3OmniModel"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":1600,"used_storage":31499212848,"created":"2026-10-19T03:38:29Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/nvidia/cosmos3-nano-libero10-sdg-export-S2_s43","likes":1,"downloads":112,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.81,"lane":"main","quant_sizes":{"FP16":{"gb":31.5,"file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-S2_s43","kind":"safetensors","community":false}},"summary":"Robotics model from nvidia (HF pipeline tag; the card has no one-line description). 15B params. License: nvidia-open-model-license.","est":{"weights":{"Q4":9.1,"Q8":16.1,"FP16":31.5},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 15.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 15.2B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-S2_s43","fmt":"safetensors","community":false}},"kv_gb":1.3,"kv_src":"rough","need":{"Q4":11.4,"Q8":18.4,"FP16":33.8},"fit":{"nv8":null,"gpu12":null,"gpu16":"Q4","nv24":"Q8","nv32":"Q8","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q4","amd24":"Q8","strix":"FP16","gpu24":"Q8","gpu32":"Q8","mac16":"Q4","mac24":"Q4","mac36":"Q8","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q8","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"estimate","amd24":"estimate","strix":"real","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"nvidia/cosmos3-nano-libero10-sdg-export-S5_2_s43","name":"cosmos3-nano-libero10-sdg-export-S5_2_s43","org":"nvidia","date":"2026-10-19","license":"nvidia-open-model-license","params":15173136576,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"robotics","modality":"robotics","summary_what":"","arch":{"class":"Cosmos3OmniModel"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":1602,"used_storage":31499212848,"created":"2026-10-19T21:19:46Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/nvidia/cosmos3-nano-libero10-sdg-export-S5_2_s43","likes":0,"downloads":179,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":1.47,"lane":"main","quant_sizes":{"FP16":{"gb":31.5,"file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-S5_2_s43","kind":"safetensors","community":false}},"summary":"Robotics model from nvidia (HF pipeline tag; the card has no one-line description). 15B params. License: nvidia-open-model-license.","est":{"weights":{"Q4":9.1,"Q8":16.1,"FP16":31.5},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 15.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 15.2B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-S5_2_s43","fmt":"safetensors","community":false}},"kv_gb":1.3,"kv_src":"rough","need":{"Q4":11.4,"Q8":18.4,"FP16":33.8},"fit":{"nv8":null,"gpu12":null,"gpu16":"Q4","nv24":"Q8","nv32":"Q8","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q4","amd24":"Q8","strix":"FP16","gpu24":"Q8","gpu32":"Q8","mac16":"Q4","mac24":"Q4","mac36":"Q8","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q8","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"estimate","amd24":"estimate","strix":"real","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"nvidia/cosmos3-nano-libero10-sdg-export-D2_s43","name":"cosmos3-nano-libero10-sdg-export-D2_s43","org":"nvidia","date":"2026-10-19","license":"nvidia-open-model-license","params":15173136576,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"robotics","modality":"robotics","summary_what":"","arch":{"class":"Cosmos3OmniModel"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":1625,"used_storage":31499212848,"created":"2026-10-19T12:33:39Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/nvidia/cosmos3-nano-libero10-sdg-export-D2_s43","likes":0,"downloads":153,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":1.17,"lane":"main","quant_sizes":{"FP16":{"gb":31.5,"file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-D2_s43","kind":"safetensors","community":false}},"summary":"Robotics model from nvidia (HF pipeline tag; the card has no one-line description). 15B params. License: nvidia-open-model-license.","est":{"weights":{"Q4":9.1,"Q8":16.1,"FP16":31.5},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 15.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 15.2B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-D2_s43","fmt":"safetensors","community":false}},"kv_gb":1.3,"kv_src":"rough","need":{"Q4":11.4,"Q8":18.4,"FP16":33.8},"fit":{"nv8":null,"gpu12":null,"gpu16":"Q4","nv24":"Q8","nv32":"Q8","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q4","amd24":"Q8","strix":"FP16","gpu24":"Q8","gpu32":"Q8","mac16":"Q4","mac24":"Q4","mac36":"Q8","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q8","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"estimate","amd24":"estimate","strix":"real","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"briaai/Fibo-2.0-dev","name":"Fibo-2.0-dev","org":"briaai","date":"2026-10-18","license":"bria-fibo","params":8127342213,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"text-to-image","modality":"image","summary_what":"Fibo 2.0 is the of Bria's FIBO family.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":18890,"used_storage":25499453513,"created":"2026-10-18T13:44:10Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/briaai/Fibo-2.0-dev","likes":137,"downloads":1024,"finetune":false,"gated":true,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"137 likes","velocity":7.56,"lane":"main","quant_sizes":{},"summary":"Fibo 2.0 is the of Bria's FIBO family. 8.1B params. License: bria-fibo.","est":{"weights":{"Q4":4.9,"Q8":8.6,"FP16":16.3},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 8.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 8.1B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 8.1B params"}},"kv_gb":0.7,"kv_src":"rough","need":{"Q4":6.6,"Q8":10.3,"FP16":18.0},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"deepseek-ai/DeepSeek-V4.1-Flash-1017","name":"DeepSeek-V4.1-Flash-1017","org":"deepseek-ai","date":"2026-10-17","license":"mit","params":763205315794,"params_src":"safetensors","active":null,"ctx":1048576,"kv_bytes_tok":81920,"dtypes":["BF16","F32","F8_E4M3","I8"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"DeepSeek-V4.1-Flash-1017 is a minor update to DeepSeek-V4.1-Flash, delivering substantial improvements across reasoning, coding, agentic tasks, and visual benchmarks.","arch":{"class":"DeepseekV41ForCausalLM","type":"MoE","experts":384,"experts_active":6},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":5118,"used_storage":510296920807,"created":"2026-10-17T02:40:38Z","first_seen":"2026-10-17T02:40:38Z","fresh":false,"url":"https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017","likes":2765,"downloads":73215,"finetune":false,"gated":false,"quants":[{"id":"mlx-community/DeepSeek-V4.1-Flash-1017-oQ2.5e-mtp","kind":"MLX"},{"id":"mlx-community/DeepSeek-V4.1-Flash-1017-oQ2.5e-ngram-mtp","kind":"MLX"},{"id":"mlx-community/DeepSeek-V4.1-Flash-1017-oQ2e-mtp","kind":"MLX"},{"id":"mlx-community/DeepSeek-V4.1-Flash-1017-oQ2e-ngram-mtp","kind":"MLX"},{"id":"mlx-community/DeepSeek-V4.1-Flash-1017-oQ3e-mtp","kind":"MLX"},{"id":"mlx-community/DeepSeek-V4.1-Flash-1017-oQ3e-ngram-mtp","kind":"MLX"},{"id":"mlx-community/DeepSeek-V4.1-Flash-1017-oQ4e-mtp","kind":"MLX"},{"id":"mlx-community/DeepSeek-V4.1-Flash-1017-oQ4e-ngram-mtp","kind":"MLX"},{"id":"unsloth/DeepSeek-V4.1-Flash-1017-GGUF","kind":"GGUF"}],"x_mentions":[{"url":"https://x.com/sfxnz/status/2112195859093794845","user":"sfxnz","likes":14},{"url":"https://x.com/the_dark_bidder/status/2112257838147854405","user":"the_dark_bidder","likes":2},{"url":"https://x.com/TSC/status/2112381585852944388","user":"TSC","likes":0}],"x_likes":832,"x_posts":42,"signal":"trusted org","velocity":822.62,"lane":"main","quant_sizes":{"Q4":{"gb":500.7,"file":"UD-Q4_K_XL","repo":"unsloth/DeepSeek-V4.1-Flash-1017-GGUF","kind":"GGUF","community":false},"Q8":{"gb":508.4,"file":"UD-Q8_K_XL","repo":"unsloth/DeepSeek-V4.1-Flash-1017-GGUF","kind":"GGUF","community":false}},"summary":"DeepSeek-V4.1-Flash-1017 is a minor update to DeepSeek-V4.1-Flash, delivering substantial improvements across reasoning, coding, agentic tasks, and visual benchmarks. Mixture-of-experts, 763B total (384 experts, 6 per token). Context: 1M tokens (config). License: mit.","ollama":"https://ollama.com/library/deepseek-v4.1-flash","est":{"weights":{"Q4":500.7,"Q8":508.4,"FP16":1526.4},"weights_src":{"Q4":{"kind":"real","file":"UD-Q4_K_XL","repo":"unsloth/DeepSeek-V4.1-Flash-1017-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"UD-Q8_K_XL","repo":"unsloth/DeepSeek-V4.1-Flash-1017-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 763.2B params"}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":502.4,"Q8":510.1,"FP16":1528.1},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":null,"amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":null,"mac512":null,"cpu32":null,"cpu64":null,"cpu128":null,"cpu256":null},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":null,"amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":null,"mac512":null,"cpu32":null,"cpu64":null,"cpu128":null,"cpu256":null},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"autotrust/AutoTrust-JEV-27B-v1.1","name":"AutoTrust-JEV-27B-v1.1","org":"autotrust","date":"2026-10-17","license":"apache-2.0","params":27356728560,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16"],"pipeline":"text-classification","modality":"text","summary_what":"AutoTrust-JEV-27B v1.1 is a text decision model that assigns probabilities to the answer options supplied with each question.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":10467,"used_storage":54727025450,"created":"2026-10-17T07:59:58Z","first_seen":"2026-10-19T11:24:28Z","fresh":false,"url":"https://huggingface.co/autotrust/AutoTrust-JEV-27B-v1.1","likes":144,"downloads":41137,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"144 likes, 41137 downloads","velocity":225.1,"lane":"main","quant_sizes":{"FP16":{"gb":54.7,"file":"official BF16 safetensors","repo":"autotrust/AutoTrust-JEV-27B-v1.1","kind":"safetensors","community":false}},"summary":"AutoTrust-JEV-27B v1.1 is a text decision model that assigns probabilities to the answer options supplied with each question. Dense, 27B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":16.4,"Q8":29.1,"FP16":54.7},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 27.4B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 27.4B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"autotrust/AutoTrust-JEV-27B-v1.1","fmt":"safetensors","community":false}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":19.5,"Q8":32.2,"FP16":57.8},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"estimate","strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"estimate","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"microsoft/Sentiment-Analysis-7B","name":"Sentiment-Analysis-7B","org":"microsoft","date":"2026-10-17","license":"mit","params":7241732096,"params_src":"safetensors","active":null,"ctx":32768,"kv_bytes_tok":131072,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"Sentiment-Analysis-7B is a 7B-parameter autoregressive language model fine-tuned for sentiment classification of user messages directed at a service, product, or agent (the \"Target\").","arch":{"class":"MistralForCausalLM","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":3448,"used_storage":14483497944,"created":"2026-10-17T00:51:18Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/microsoft/Sentiment-Analysis-7B","likes":22,"downloads":661,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":3.58,"lane":"main","quant_sizes":{"Q4":{"gb":4.4,"file":"Q4_K_M","repo":"mradermacher/Sentiment-Analysis-7B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":7.7,"file":"Q8_0","repo":"mradermacher/Sentiment-Analysis-7B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":14.5,"file":"F16","repo":"mradermacher/Sentiment-Analysis-7B-GGUF","kind":"GGUF","community":false}},"summary":"Sentiment-Analysis-7B is a 7B-parameter autoregressive language model fine-tuned for sentiment classification of user messages directed at a service, product, or agent (the \"Target\"). Dense, 7.2B params. Context: 32K tokens (config). License: mit.","est":{"weights":{"Q4":4.4,"Q8":7.7,"FP16":14.5},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/Sentiment-Analysis-7B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/Sentiment-Analysis-7B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"mradermacher/Sentiment-Analysis-7B-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":6.5,"Q8":9.8,"FP16":16.6},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"Comfy-Org/MOSS-Image","name":"MOSS-Image","org":"Comfy-Org","date":"2026-10-16","license":"apache-2.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":3746,"used_storage":68453530480,"created":"2026-10-16T12:42:32Z","first_seen":"2026-10-16T12:42:32Z","fresh":false,"url":"https://huggingface.co/Comfy-Org/MOSS-Image","likes":424,"downloads":1944278,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"424 likes, 1944278 downloads","velocity":9594.81,"lane":"main","quant_sizes":{},"summary":"License: apache-2.0.","est":null,"license_class":"permissive"},{"type":"model","id":"lodestones/MOSS-Image-SFT-8-steps-dmd","name":"MOSS-Image-SFT-8-steps-dmd","org":"lodestones","date":"2026-10-16","license":"apache-2.0","params":5082308624,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"","modality":"","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":2650,"used_storage":137927743472,"created":"2026-10-16T10:28:56Z","first_seen":"2026-10-16T10:28:56Z","fresh":false,"url":"https://huggingface.co/lodestones/MOSS-Image-SFT-8-steps-dmd","likes":209,"downloads":113000,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"209 likes, 113000 downloads","velocity":552.48,"lane":"main","quant_sizes":{},"summary":"5.1B params. License: apache-2.0.","est":{"weights":{"Q4":3.0,"Q8":5.4,"FP16":10.2},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 5.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 5.1B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 5.1B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":4.5,"Q8":6.9,"FP16":11.7},"fit":{"nv8":"Q8","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"nvidia/NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16","name":"NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16","org":"nvidia","date":"2026-10-16","license":"openmdw-1.1","params":3975229440,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":294912,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16 is a looped (latent-recurrent) large language model (LLM) trained by NVIDIA.","arch":{"class":"DoryForCausalLM","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19677,"used_storage":7969108053,"created":"2026-10-16T18:25:29Z","first_seen":"2026-10-16T18:25:29Z","fresh":false,"url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16","likes":181,"downloads":4576,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":24.15,"lane":"main","quant_sizes":{"Q4":{"gb":2.9,"file":"Q4_K_XL","repo":"cloudnathan5/Dory1.0-4B-GGUF","kind":"GGUF","community":true},"Q8":{"gb":4.2,"file":"Q8_0","repo":"cloudnathan5/Dory1.0-4B-GGUF","kind":"GGUF","community":true},"FP16":{"gb":8.0,"file":"BF16","repo":"cloudnathan5/Dory1.0-4B-GGUF","kind":"GGUF","community":true}},"summary":"NVIDIA-Nemotron-Labs-Dory1.0-4B-BF16 is a looped (latent-recurrent) large language model (LLM) trained by NVIDIA. Dense, 4B params. Context: 256K tokens (config). License: openmdw-1.1.","ollama":"https://ollama.com/library/nemotron","est":{"weights":{"Q4":2.9,"Q8":4.2,"FP16":8.0},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_XL","repo":"cloudnathan5/Dory1.0-4B-GGUF","fmt":"GGUF","community":true},"Q8":{"kind":"real","file":"Q8_0","repo":"cloudnathan5/Dory1.0-4B-GGUF","fmt":"GGUF","community":true},"FP16":{"kind":"real","file":"BF16","repo":"cloudnathan5/Dory1.0-4B-GGUF","fmt":"GGUF","community":true}},"kv_gb":2.4,"kv_src":"config","need":{"Q4":6.3,"Q8":7.6,"FP16":11.4},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"ornith-ai/Ornith-1.5-122B","name":"Ornith-1.5-122B","org":"ornith-ai","date":"2026-10-16","license":"mit","params":125086497008,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":98304,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"Qwen3_5MoeForConditionalGeneration","type":"MoE","experts":256,"experts_active":8},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19839,"used_storage":250188099758,"created":"2026-10-16T07:14:27Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/ornith-ai/Ornith-1.5-122B","likes":157,"downloads":30407,"finetune":false,"gated":false,"quants":[{"id":"bartowski/Ornith-1.5-122B-GGUF","kind":"GGUF"},{"id":"unsloth/Ornith-1.5-122B-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"157 likes, 30407 downloads","velocity":146.84,"lane":"main","quant_sizes":{"Q4":{"gb":75.8,"file":"Q4_K_M","repo":"ornith-ai/Ornith-1.5-122B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":132.6,"file":"Q8_0","repo":"ornith-ai/Ornith-1.5-122B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":249.4,"file":"BF16","repo":"ornith-ai/Ornith-1.5-122B-GGUF","kind":"GGUF","community":false}},"summary":"Text generation model from ornith-ai (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 125B total (256 experts, 8 per token). Context: 256K tokens (config). License: mit.","ollama":"https://ollama.com/library/ornith-1.5","est":{"weights":{"Q4":75.8,"Q8":132.6,"FP16":249.4},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"ornith-ai/Ornith-1.5-122B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"ornith-ai/Ornith-1.5-122B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"ornith-ai/Ornith-1.5-122B-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.8,"kv_src":"config","need":{"Q4":77.6,"Q8":134.4,"FP16":251.2},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":"Q4","spark1":"Q4","spark2":"Q8","amd16":null,"amd24":null,"strix":"Q4","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":"Q4","mac192":"Q8","mac256":"Q8","mac512":"FP16","cpu32":null,"cpu64":null,"cpu128":"Q4","cpu256":"Q8"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":null,"cpu64":null,"cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"tencent/Qwen3-8B-DSpine-b16","name":"Qwen3-8B-DSpine-b16","org":"tencent","date":"2026-10-16","license":"apache-2.0","params":1237763840,"params_src":"safetensors","active":null,"ctx":40960,"kv_bytes_tok":20480,"dtypes":["BF16"],"pipeline":"","modality":"","summary_what":"DSpine is a parallel drafter for speculative decoding.","arch":{"class":"DFlashDraftModel","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":3829,"used_storage":2475535344,"created":"2026-10-16T14:52:38Z","first_seen":"2026-10-16T14:52:38Z","fresh":false,"url":"https://huggingface.co/tencent/Qwen3-8B-DSpine-b16","likes":25,"downloads":558,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":2.91,"lane":"main","quant_sizes":{"FP16":{"gb":2.5,"file":"official BF16 safetensors","repo":"tencent/Qwen3-8B-DSpine-b16","kind":"safetensors","community":false}},"summary":"DSpine is a parallel drafter for speculative decoding. Dense, 1.2B params. Context: 40K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/qwen3.8","est":{"weights":{"Q4":0.7,"Q8":1.3,"FP16":2.5},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 1.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 1.2B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"tencent/Qwen3-8B-DSpine-b16","fmt":"safetensors","community":false}},"kv_gb":0.2,"kv_src":"config","need":{"Q4":1.9,"Q8":2.5,"FP16":3.7},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"nvidia/Muse-Spark-Open-816B-A42B-DFlash2","name":"Muse-Spark-Open-816B-A42B-DFlash2","org":"nvidia","date":"2026-10-16","license":"nvidia-open-model-agreement-license","params":4384162432,"params_src":"safetensors","active":null,"ctx":1048576,"kv_bytes_tok":20480,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"","arch":{"class":"DFlash2DraftModel","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":17328,"used_storage":8768333792,"created":"2026-10-16T04:40:26Z","first_seen":"2026-10-16T04:40:26Z","fresh":false,"url":"https://huggingface.co/nvidia/Muse-Spark-Open-816B-A42B-DFlash2","likes":8,"downloads":1203,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":5.75,"lane":"main","quant_sizes":{"FP16":{"gb":8.8,"file":"official BF16 safetensors","repo":"nvidia/Muse-Spark-Open-816B-A42B-DFlash2","kind":"safetensors","community":false}},"summary":"Image text to text model from nvidia (HF pipeline tag; the card has no one-line description). Dense, 4.4B params. Context: 1M tokens (config). License: nvidia-open-model-agreement-license.","ollama":"https://ollama.com/library/muse-spark-open","est":{"weights":{"Q4":2.6,"Q8":4.7,"FP16":8.8},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 4.4B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 4.4B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"nvidia/Muse-Spark-Open-816B-A42B-DFlash2","fmt":"safetensors","community":false}},"kv_gb":0.2,"kv_src":"config","need":{"Q4":3.8,"Q8":5.9,"FP16":10.0},"fit":{"nv8":"Q8","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"microsoft/scFM-dataselection","name":"scFM-dataselection","org":"microsoft","date":"2026-10-16","license":"mit","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":5074,"used_storage":36491689646,"created":"2026-10-16T13:54:04Z","first_seen":"2026-10-16T13:54:04Z","fresh":false,"url":"https://huggingface.co/microsoft/scFM-dataselection","likes":3,"downloads":0,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.01,"lane":"main","quant_sizes":{},"summary":"License: mit.","est":null,"license_class":"permissive"},{"type":"model","id":"leemysw/yovoice-model","name":"yovoice-model","org":"leemysw","date":"2026-10-16","license":"per-model","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"text-to-speech","modality":"audio","summary_what":"\u4e00\u822c\u4e0d\u9700\u8981\u624b\u52a8\u4e0b\u8f7d\uff1a\u5728 yovoice \u7684 \u8bbe\u7f6e \u203a \u6a21\u578b \u4e2d\u4e0b\u8f7d\uff08\u6216\u8fd0\u884c yovoice models download \uff09\uff0c\u5e94\u7528\u4f1a\u81ea\u52a8\u6821\u9a8c\u6587\u4ef6\u5927\u5c0f\u548c SHA256\u3002","arch":{},"claims":[],"has_safetensors":false,"has_weights":true,"card_len":3711,"used_storage":503252440,"created":"2026-10-16T08:32:23Z","first_seen":"2026-10-16T08:32:23Z","fresh":false,"url":"https://huggingface.co/leemysw/yovoice-model","likes":0,"downloads":83944,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"83944 downloads","velocity":405.82,"lane":"main","quant_sizes":{},"summary":"\u4e00\u822c\u4e0d\u9700\u8981\u624b\u52a8\u4e0b\u8f7d\uff1a\u5728 yovoice \u7684 \u8bbe\u7f6e \u203a \u6a21\u578b \u4e2d\u4e0b\u8f7d\uff08\u6216\u8fd0\u884c yovoice models download \uff09\uff0c\u5e94\u7528\u4f1a\u81ea\u52a8\u6821\u9a8c\u6587\u4ef6\u5927\u5c0f\u548c SHA256\u3002. License: per-model.","est":null,"license_class":"restricted"},{"type":"model","id":"tencent/Hy-Memory-30B-A3B","name":"Hy-Memory-30B-A3B","org":"tencent","date":"2026-10-15","license":"apache-2.0","params":30064725888,"params_src":"safetensors","active":3000000000.0,"ctx":262144,"kv_bytes_tok":98304,"dtypes":["BF16","F32"],"pipeline":"text-generation","modality":"text","summary_what":"Hy-Memory-30B-A3B is a compact long-term-memory writer model (~3B active parameters) developed by Tencent Hunyuan to execute the write path inside external memory frameworks.","arch":{"class":"HYV3ForCausalLM","type":"MoE","experts":128,"experts_active":8,"active":3000000000.0,"active_src":"name"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":9066,"used_storage":60132268315,"created":"2026-10-15T09:52:00Z","first_seen":"2026-10-15T09:52:00Z","fresh":false,"url":"https://huggingface.co/tencent/Hy-Memory-30B-A3B","likes":310,"downloads":3923,"finetune":false,"gated":false,"quants":[{"id":"mlx-community/Hy-Memory-30B-A3B-4bit","kind":"MLX"},{"id":"mlx-community/Hy-Memory-30B-A3B-8bit","kind":"MLX"},{"id":"tencent/Hy-Memory-30B-A3B-FP8","kind":"quant"},{"id":"tencent/Hy-Memory-30B-A3B-GGUF","kind":"GGUF"}],"x_mentions":[{"url":"https://x.com/the_rza_/status/2112894728173686935","user":"the_rza_","likes":0}],"x_likes":0,"x_posts":1,"signal":"trusted org","velocity":28.44,"lane":"main","quant_sizes":{"Q4":{"gb":18.2,"file":"Q4_K_M","repo":"tencent/Hy-Memory-30B-A3B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":32.0,"file":"MLX 8-bit","repo":"mlx-community/Hy-Memory-30B-A3B-8bit","kind":"MLX","community":false},"FP16":{"gb":60.1,"file":"official BF16 safetensors","repo":"tencent/Hy-Memory-30B-A3B","kind":"safetensors","community":false}},"summary":"Hy-Memory-30B-A3B is a compact long-term-memory writer model (~3B active parameters) developed by Tencent Hunyuan to execute the write path inside external memory frameworks. Mixture-of-experts, 30B total, 3B active (128 experts, 8 per token). Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":18.2,"Q8":32.0,"FP16":60.1},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"tencent/Hy-Memory-30B-A3B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"MLX 8-bit","repo":"mlx-community/Hy-Memory-30B-A3B-8bit","fmt":"MLX","community":false},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"tencent/Hy-Memory-30B-A3B","fmt":"safetensors","community":false}},"kv_gb":0.8,"kv_src":"config","need":{"Q4":20.0,"Q8":33.8,"FP16":61.9},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"InternScience/Sage-A1.5","name":"Sage-A1.5","org":"InternScience","date":"2026-10-15","license":"apache-2.0","params":35951822704,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":81920,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"Sage\u2011A1.5 is a 35B Mixture\u2011of\u2011Experts agentic model based on Qwen3.6 and Agents-A1 from Shanghai AI Laboratory, built to advance long-horizon scientific problem solving through adaptive capability composition.","arch":{"class":"Qwen3_5MoeForConditionalGeneration","type":"MoE","experts":256,"experts_active":8},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19836,"used_storage":71928317664,"created":"2026-10-15T08:34:39Z","first_seen":"2026-10-21T11:18:13Z","fresh":false,"url":"https://huggingface.co/InternScience/Sage-A1.5","likes":267,"downloads":2033,"finetune":false,"gated":false,"quants":[{"id":"InternScience/Sage-A1.5-F16-GGUF","kind":"GGUF"},{"id":"InternScience/Sage-A1.5-FP8","kind":"quant"},{"id":"InternScience/Sage-A1.5-Q4_K_M-GGUF","kind":"GGUF"},{"id":"InternScience/Sage-A1.5-Q8_0-GGUF","kind":"GGUF"},{"id":"mlx-community/Sage-A1.5-OptiQ-4bit","kind":"MLX"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"267 likes","velocity":9.96,"lane":"main","quant_sizes":{"Q4":{"gb":21.7,"file":"Q4_K_M","repo":"InternScience/Sage-A1.5-Q4_K_M-GGUF","kind":"GGUF","community":false},"FP16":{"gb":71.1,"file":"F16","repo":"InternScience/Sage-A1.5-F16-GGUF","kind":"GGUF","community":false},"Q8":{"gb":37.8,"file":"Q8_0","repo":"InternScience/Sage-A1.5-Q8_0-GGUF","kind":"GGUF","community":false}},"summary":"Sage\u2011A1.5 is a 35B Mixture\u2011of\u2011Experts agentic model based on Qwen3.6 and Agents-A1 from Shanghai AI Laboratory, built to advance long-horizon scientific problem solving through adaptive capability composition. Mixture-of-experts, 36B total (256 experts, 8 per token). Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":21.7,"Q8":37.8,"FP16":71.1},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"InternScience/Sage-A1.5-Q4_K_M-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"InternScience/Sage-A1.5-Q8_0-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"InternScience/Sage-A1.5-F16-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":23.4,"Q8":39.5,"FP16":72.8},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":null,"strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q4","mac64":"Q8","mac96":"Q8","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"perplexity-ai/pplx-decider-v1.2","name":"pplx-decider-v1.2","org":"perplexity-ai","date":"2026-10-15","license":"apache-2.0","params":26085330160,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16"],"pipeline":"text-classification","modality":"text","summary_what":"pplx-decider-v1.2 is the next version of our pplx-decider-v1.1-27b decision model, on the same Qwen3.8-27B backbone.","arch":{"class":"Qwen3_5Model","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19021,"used_storage":52193403540,"created":"2026-10-15T07:40:00Z","first_seen":"2026-10-15T07:40:00Z","fresh":false,"url":"https://huggingface.co/perplexity-ai/pplx-decider-v1.2","likes":44,"downloads":1627,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":7.21,"lane":"main","quant_sizes":{"FP16":{"gb":52.2,"file":"official BF16 safetensors","repo":"perplexity-ai/pplx-decider-v1.2","kind":"safetensors","community":false}},"summary":"pplx-decider-v1.2 is the next version of our pplx-decider-v1.1-27b decision model, on the same Qwen3.8-27B backbone. Dense, 26B params. Context: 256K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/pplx-decider","est":{"weights":{"Q4":15.7,"Q8":27.7,"FP16":52.2},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 26.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 26.1B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"perplexity-ai/pplx-decider-v1.2","fmt":"safetensors","community":false}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":18.8,"Q8":30.8,"FP16":55.3},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q8","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":"Q8","mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"estimate","strix":"real","gpu24":null,"gpu32":"estimate","mac16":null,"mac24":null,"mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"nvidia/Nemotron-3.5-Rerank-8B-BF16","name":"Nemotron-3.5-Rerank-8B-BF16","org":"nvidia","date":"2026-10-15","license":"openmdw-1.1","params":8381159424,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":139264,"dtypes":["BF16"],"pipeline":"text-ranking","modality":"embedding","summary_what":"","arch":{"class":"Mistral3ForSequenceClassification","type":"dense"},"claims":["Nemotron-3.5-Rerank-8B-BF16 achieves state-of-the-art results on the new ViDoRe v3.1 Reranking leaderboard) as of October 20, 2026."],"has_safetensors":true,"has_weights":true,"card_len":19375,"used_storage":16779485075,"created":"2026-10-15T08:42:42Z","first_seen":"2026-10-15T08:42:42Z","fresh":false,"url":"https://huggingface.co/nvidia/Nemotron-3.5-Rerank-8B-BF16","likes":39,"downloads":2019,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":8.92,"lane":"main","quant_sizes":{"FP16":{"gb":16.8,"file":"official BF16 safetensors","repo":"nvidia/Nemotron-3.5-Rerank-8B-BF16","kind":"safetensors","community":false}},"summary":"Text ranking model from nvidia (HF pipeline tag; the card has no one-line description). Dense, 8.4B params. Context: 256K tokens (config). License: openmdw-1.1. Vendor-reported: \"Nemotron-3.5-Rerank-8B-BF16 achieves state-of-the-art results on the new ViDoRe v3.1 Reranking leaderboard) as of October 20, 2026.\"","ollama":"https://ollama.com/library/nemotron3","est":{"weights":{"Q4":5.0,"Q8":8.9,"FP16":16.8},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 8.4B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 8.4B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"nvidia/Nemotron-3.5-Rerank-8B-BF16","fmt":"safetensors","community":false}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":7.1,"Q8":11.0,"FP16":18.9},"fit":{"nv8":null,"gpu12":"Q8","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"Q8","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":"estimate","gpu16":"estimate","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"estimate","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"estimate","mac24":"estimate","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"perplexity-ai/pplx-decider-flash-v1.2","name":"pplx-decider-flash-v1.2","org":"perplexity-ai","date":"2026-10-15","license":"apache-2.0","params":4539265536,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":131072,"dtypes":["BF16"],"pipeline":"text-classification","modality":"text","summary_what":"pplx-decider-flash-v1.2 is the small version of pplx-decider-v1.2: the same joint request format, the same API and the same text and image support, at about one seventh of the size.","arch":{"class":"Qwen3_5Model","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":12281,"used_storage":9099911300,"created":"2026-10-15T16:57:38Z","first_seen":"2026-10-15T16:57:38Z","fresh":false,"url":"https://huggingface.co/perplexity-ai/pplx-decider-flash-v1.2","likes":39,"downloads":1796,"finetune":true,"gated":false,"quants":[{"id":"perplexity-ai/pplx-decider-flash-gguf-v1.2","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":8.25,"lane":"main","quant_sizes":{"Q8":{"gb":5.2,"file":"Q8_0","repo":"perplexity-ai/pplx-decider-flash-gguf-v1.2","kind":"GGUF","community":false},"FP16":{"gb":9.1,"file":"official BF16 safetensors","repo":"perplexity-ai/pplx-decider-flash-v1.2","kind":"safetensors","community":false}},"summary":"pplx-decider-flash-v1.2 is the small version of pplx-decider-v1.2: the same joint request format, the same API and the same text and image support, at about one seventh of the size. Dense, 4.5B params. Context: 256K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/pplx-decider","est":{"weights":{"Q4":2.7,"Q8":5.2,"FP16":9.1},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 4.5B params"},"Q8":{"kind":"real","file":"Q8_0","repo":"perplexity-ai/pplx-decider-flash-gguf-v1.2","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"perplexity-ai/pplx-decider-flash-v1.2","fmt":"safetensors","community":false}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":4.8,"Q8":7.3,"FP16":11.2},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"nvidia/RFD4-Proteina-prep","name":"RFD4-Proteina-prep","org":"nvidia","date":"2026-10-15","license":"bsd-3-clause","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"","arch":{},"claims":[],"has_safetensors":false,"has_weights":false,"card_len":14362,"used_storage":0,"created":"2026-10-15T15:57:21Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/nvidia/RFD4-Proteina-prep","likes":0,"downloads":40,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.18,"lane":"main","quant_sizes":{},"summary":"License: bsd-3-clause.","est":null,"license_class":"permissive"},{"type":"model","id":"nvidia/RFD4-Proteina-Diffusers","name":"RFD4-Proteina-Diffusers","org":"nvidia","date":"2026-10-15","license":"bsd-3-clause","params":178448304,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"","modality":"","summary_what":"Use the model for computational biomolecular design and follow designs with appropriate structural and experimental evaluation.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":13134,"used_storage":9119289987,"created":"2026-10-15T03:07:08Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/nvidia/RFD4-Proteina-Diffusers","likes":0,"downloads":147,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.62,"lane":"main","quant_sizes":{},"summary":"Use the model for computational biomolecular design and follow designs with appropriate structural and experimental evaluation. 178M params. License: bsd-3-clause.","est":{"weights":{"Q4":0.1,"Q8":0.2,"FP16":0.4},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.2B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.2B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.6,"Q8":1.7,"FP16":1.9},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"black-forest-labs/flux-3-image-klein-comfy","name":"flux-3-image-klein-comfy","org":"black-forest-labs","date":"2026-10-14","license":"flux-kommunity-license","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"image-to-image","modality":"image","summary_what":"FLUX 3 Image Klein is a 5.6B open weights model for text-to-image generation and image editing.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":11492,"used_storage":93726613295,"created":"2026-10-14T12:00:13Z","first_seen":"2026-10-14T12:00:13Z","fresh":false,"url":"https://huggingface.co/black-forest-labs/flux-3-image-klein-comfy","likes":1495,"downloads":590742,"finetune":true,"gated":true,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":2355.87,"lane":"main","quant_sizes":{"Q4":{"gb":6.8,"file":"Q4_K_M","repo":"realrebelai/FLUX-3-Image-Klein_GGUFs","kind":"GGUF","community":true},"Q8":{"gb":12.1,"file":"Q8_0","repo":"realrebelai/FLUX-3-Image-Klein_GGUFs","kind":"GGUF","community":true}},"summary":"FLUX 3 Image Klein is a 5.6B open weights model for text-to-image generation and image editing. License: flux-kommunity-license.","est":{"weights":{"Q4":6.8,"Q8":12.1},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"realrebelai/FLUX-3-Image-Klein_GGUFs","fmt":"GGUF","community":true},"Q8":{"kind":"real","file":"Q8_0","repo":"realrebelai/FLUX-3-Image-Klein_GGUFs","fmt":"GGUF","community":true}},"kv_gb":1.0,"kv_src":"rough","need":{"Q4":8.8,"Q8":14.1},"fit":{"nv8":null,"gpu12":"Q4","gpu16":"Q8","nv24":"Q8","nv32":"Q8","nv48":"Q8","nv96":"Q8","spark1":"Q8","spark2":"Q8","amd16":"Q8","amd24":"Q8","strix":"Q8","gpu24":"Q8","gpu32":"Q8","mac16":"Q4","mac24":"Q8","mac36":"Q8","mac48":"Q8","mac64":"Q8","mac96":"Q8","mac128":"Q8","mac192":"Q8","mac256":"Q8","mac512":"Q8","cpu32":"Q8","cpu64":"Q8","cpu128":"Q8","cpu256":"Q8"},"fit_src":{"nv8":null,"gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"restricted"},{"type":"model","id":"mistralai/Timestral-209M","name":"Timestral-209M","org":"mistralai","date":"2026-10-14","license":"other","params":209312256,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"time-series-forecasting","modality":"other","summary_what":"Timestral is a family of time series foundation models for zero-shot forecasting, available in 5M, 28M, 123M, and 209M parameter versions.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":5341,"used_storage":837271648,"created":"2026-10-14T08:52:36Z","first_seen":"2026-10-14T08:52:36Z","fresh":false,"url":"https://huggingface.co/mistralai/Timestral-209M","likes":338,"downloads":8731,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":35.63,"lane":"main","quant_sizes":{},"summary":"Timestral is a family of time series foundation models for zero-shot forecasting, available in 5M, 28M, 123M, and 209M parameter versions. 209M params. License: other.","est":{"weights":{"Q4":0.1,"Q8":0.2,"FP16":0.4},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.2B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.2B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.6,"Q8":1.7,"FP16":1.9},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"mistralai/Timestral-5M","name":"Timestral-5M","org":"mistralai","date":"2026-10-14","license":"other","params":5208960,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"time-series-forecasting","modality":"other","summary_what":"Timestral is a family of time series foundation models for zero-shot forecasting, available in 5M, 28M, 123M, and 209M parameter versions.","arch":{},"claims":["All Timestreal models outperform every other model at the respective model size."],"has_safetensors":true,"has_weights":true,"card_len":3192,"used_storage":20843960,"created":"2026-10-14T13:41:13Z","first_seen":"2026-10-14T13:41:13Z","fresh":false,"url":"https://huggingface.co/mistralai/Timestral-5M","likes":72,"downloads":4735,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":19.25,"lane":"main","quant_sizes":{},"summary":"Timestral is a family of time series foundation models for zero-shot forecasting, available in 5M, 28M, 123M, and 209M parameter versions. 5M params. License: other. Vendor-reported: \"All Timestreal models outperform every other model at the respective model size.\"","est":{"weights":{"Q4":0.0,"Q8":0.0,"FP16":0.0},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.0B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.0B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.0B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.5,"Q8":1.5,"FP16":1.5},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"black-forest-labs/flux-3-image-klein-pup","name":"flux-3-image-klein-pup","org":"black-forest-labs","date":"2026-10-14","license":"apache-2.0","params":9409813744,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":131072,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"FLUX 3 Image Klein Prompt Upsampler is a 9B vision-language model that expands a short image request into the detailed prompt FLUX 3 Image Klein works best with.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":6969,"used_storage":18839711717,"created":"2026-10-14T20:40:29Z","first_seen":"2026-10-14T20:40:29Z","fresh":false,"url":"https://huggingface.co/black-forest-labs/flux-3-image-klein-pup","likes":67,"downloads":47164,"finetune":true,"gated":false,"quants":[{"id":"mlx-community/flux-3-image-klein-pup-4bit","kind":"MLX"},{"id":"mlx-community/flux-3-image-klein-pup-8bit","kind":"MLX"},{"id":"unsloth/flux-3-image-klein-pup-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":194.59,"lane":"main","quant_sizes":{"Q4":{"gb":5.6,"file":"Q4_K_M","repo":"unsloth/flux-3-image-klein-pup-GGUF","kind":"GGUF","community":false},"Q8":{"gb":9.5,"file":"Q8_0","repo":"unsloth/flux-3-image-klein-pup-GGUF","kind":"GGUF","community":false},"FP16":{"gb":17.9,"file":"BF16","repo":"unsloth/flux-3-image-klein-pup-GGUF","kind":"GGUF","community":false}},"summary":"FLUX 3 Image Klein Prompt Upsampler is a 9B vision-language model that expands a short image request into the detailed prompt FLUX 3 Image Klein works best with. Dense, 9.4B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":5.6,"Q8":9.5,"FP16":17.9},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"unsloth/flux-3-image-klein-pup-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"unsloth/flux-3-image-klein-pup-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"unsloth/flux-3-image-klein-pup-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":7.7,"Q8":11.6,"FP16":20.0},"fit":{"nv8":null,"gpu12":"Q4","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"Q8","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"ibm-granite/granite-switch-4.1-8b","name":"granite-switch-4.1-8b","org":"ibm-granite","date":"2026-10-14","license":"apache-2.0","params":8776400896,"params_src":"safetensors","active":null,"ctx":131072,"kv_bytes_tok":172032,"dtypes":["I64","F32","BF16"],"pipeline":"text-generation","modality":"text","summary_what":"Granite Switch 4.1 8B is a modular LLM that composes a base model, multiple adapters and an encoder together so that a single hosted endpoint can move between natural-language responses, general and specialized decisions, and speech-to-text.","arch":{"class":"GraniteSwitchForCausalLM","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19896,"used_storage":17554232752,"created":"2026-10-14T21:38:09Z","first_seen":"2026-10-14T21:38:09Z","fresh":false,"url":"https://huggingface.co/ibm-granite/granite-switch-4.1-8b","likes":43,"downloads":2394,"finetune":false,"gated":false,"quants":[{"id":"ibm-granite/granite-switch-4.1-8b-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":10.08,"lane":"main","quant_sizes":{"Q4":{"gb":6.4,"file":"Q4_K_M","repo":"ibm-granite/granite-switch-4.1-8b-GGUF","kind":"GGUF","community":false},"Q8":{"gb":10.1,"file":"Q8_0","repo":"ibm-granite/granite-switch-4.1-8b-GGUF","kind":"GGUF","community":false},"FP16":{"gb":18.0,"file":"BF16","repo":"ibm-granite/granite-switch-4.1-8b-GGUF","kind":"GGUF","community":false}},"summary":"Granite Switch 4.1 8B is a modular LLM that composes a base model, multiple adapters and an encoder together so that a single hosted endpoint can move between natural-language responses, general and specialized decisions, and speech-to-text. Dense, 8.8B params. Context: 128K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":6.4,"Q8":10.1,"FP16":18.0},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"ibm-granite/granite-switch-4.1-8b-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"ibm-granite/granite-switch-4.1-8b-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"ibm-granite/granite-switch-4.1-8b-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.4,"kv_src":"config","need":{"Q4":8.8,"Q8":12.5,"FP16":20.4},"fit":{"nv8":null,"gpu12":"Q4","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q4","mac24":"Q8","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"inclusionAI/Ming-3.0-asr-tiny","name":"Ming-3.0-asr-tiny","org":"inclusionAI","date":"2026-10-14","license":"mit","params":8865800992,"params_src":"safetensors","active":null,"ctx":8192,"kv_bytes_tok":27648,"dtypes":["F32","BF16"],"pipeline":"automatic-speech-recognition","modality":"audio","summary_what":"","arch":{"class":"MingV3ASRForConditionalGeneration","type":"MoE","experts":128,"experts_active":8},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":7228,"used_storage":36821408265,"created":"2026-10-14T09:27:09Z","first_seen":"2026-10-19T23:17:53Z","fresh":false,"url":"https://huggingface.co/inclusionAI/Ming-3.0-asr-tiny","likes":42,"downloads":564,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":2.39,"lane":"main","quant_sizes":{"FP16":{"gb":17.7,"file":"official BF16 safetensors","repo":"inclusionAI/Ming-3.0-asr-tiny","kind":"safetensors","community":false}},"summary":"Automatic speech recognition model from inclusionAI (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 8.9B total (128 experts, 8 per token). Context: 8K tokens (config). License: mit.","est":{"weights":{"Q4":5.3,"Q8":9.4,"FP16":17.7},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 8.9B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 8.9B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"inclusionAI/Ming-3.0-asr-tiny","fmt":"safetensors","community":false}},"kv_gb":0.2,"kv_src":"config","need":{"Q4":6.5,"Q8":10.6,"FP16":18.9},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"Q8","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"estimate","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"estimate","mac24":"estimate","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"mistralai/Timestral-123M","name":"Timestral-123M","org":"mistralai","date":"2026-10-14","license":"other","params":122606720,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"time-series-forecasting","modality":"other","summary_what":"Timestral is a family of time series foundation models for zero-shot forecasting, available in 5M, 28M, 123M, and 209M parameter versions.","arch":{},"claims":["All Timestreal models outperform every other model at the respective model size."],"has_safetensors":true,"has_weights":true,"card_len":3200,"used_storage":490445904,"created":"2026-10-14T17:51:04Z","first_seen":"2026-10-14T17:51:04Z","fresh":false,"url":"https://huggingface.co/mistralai/Timestral-123M","likes":41,"downloads":3014,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":12.44,"lane":"main","quant_sizes":{},"summary":"Timestral is a family of time series foundation models for zero-shot forecasting, available in 5M, 28M, 123M, and 209M parameter versions. 123M params. License: other. Vendor-reported: \"All Timestreal models outperform every other model at the respective model size.\"","est":{"weights":{"Q4":0.1,"Q8":0.1,"FP16":0.2},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.1B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.1B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.6,"Q8":1.6,"FP16":1.7},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"mistralai/Timestral-28M","name":"Timestral-28M","org":"mistralai","date":"2026-10-14","license":"other","params":27821184,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"time-series-forecasting","modality":"other","summary_what":"Timestral is a family of time series foundation models for zero-shot forecasting, available in 5M, 28M, 123M, and 209M parameter versions.","arch":{},"claims":["All Timestreal models outperform every other model at the respective model size."],"has_safetensors":true,"has_weights":true,"card_len":3197,"used_storage":111296464,"created":"2026-10-14T17:50:53Z","first_seen":"2026-10-14T17:50:53Z","fresh":false,"url":"https://huggingface.co/mistralai/Timestral-28M","likes":38,"downloads":5811,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":23.82,"lane":"main","quant_sizes":{},"summary":"Timestral is a family of time series foundation models for zero-shot forecasting, available in 5M, 28M, 123M, and 209M parameter versions. 28M params. License: other. Vendor-reported: \"All Timestreal models outperform every other model at the respective model size.\"","est":{"weights":{"Q4":0.0,"Q8":0.0,"FP16":0.1},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.0B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.0B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.0B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.5,"Q8":1.5,"FP16":1.6},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"ibm-granite/granite-switch-4.1-30b","name":"granite-switch-4.1-30b","org":"ibm-granite","date":"2026-10-14","license":"apache-2.0","params":29989814272,"params_src":"safetensors","active":null,"ctx":131072,"kv_bytes_tok":270336,"dtypes":["I64","F32","BF16"],"pipeline":"text-generation","modality":"text","summary_what":"Granite Switch 4.1 30B is a modular LLM that composes a base model, multiple adapters and an encoder together so that a single hosted endpoint can move between natural-language responses, general and specialized decisions, and speech-to-text.","arch":{"class":"GraniteSwitchForCausalLM","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19895,"used_storage":59981121744,"created":"2026-10-14T21:23:12Z","first_seen":"2026-10-14T21:23:12Z","fresh":false,"url":"https://huggingface.co/ibm-granite/granite-switch-4.1-30b","likes":29,"downloads":1659,"finetune":false,"gated":false,"quants":[{"id":"ibm-granite/granite-switch-4.1-30b-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":6.98,"lane":"main","quant_sizes":{"Q4":{"gb":21.0,"file":"Q4_K_M","repo":"ibm-granite/granite-switch-4.1-30b-GGUF","kind":"GGUF","community":false},"Q8":{"gb":34.0,"file":"Q8_0","repo":"ibm-granite/granite-switch-4.1-30b-GGUF","kind":"GGUF","community":false},"FP16":{"gb":61.1,"file":"BF16","repo":"ibm-granite/granite-switch-4.1-30b-GGUF","kind":"GGUF","community":false}},"summary":"Granite Switch 4.1 30B is a modular LLM that composes a base model, multiple adapters and an encoder together so that a single hosted endpoint can move between natural-language responses, general and specialized decisions, and speech-to-text. Dense, 30B params. Context: 128K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":21.0,"Q8":34.0,"FP16":61.1},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"ibm-granite/granite-switch-4.1-30b-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"ibm-granite/granite-switch-4.1-30b-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"ibm-granite/granite-switch-4.1-30b-GGUF","fmt":"GGUF","community":false}},"kv_gb":2.2,"kv_src":"config","need":{"Q4":24.2,"Q8":37.2,"FP16":64.3},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":null,"strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q4","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"microsoft/UserRepGen-4B","name":"UserRepGen-4B","org":"microsoft","date":"2026-10-14","license":"mit","params":4539265536,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":131072,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"UserRepGen-4B is a fine-tuned Qwen3.5-4B language model that generates a structured, natural-language user representation from a multi-turn user-assistant conversation.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19804,"used_storage":9091873628,"created":"2026-10-14T20:30:00Z","first_seen":"2026-10-14T20:30:00Z","fresh":false,"url":"https://huggingface.co/microsoft/UserRepGen-4B","likes":24,"downloads":1630,"finetune":true,"gated":false,"quants":[{"id":"bartowski/UserRepGen-4B-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":6.81,"lane":"main","quant_sizes":{"Q4":{"gb":2.7,"file":"Q4_K_M","repo":"bartowski/UserRepGen-4B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":4.5,"file":"Q8_0","repo":"bartowski/UserRepGen-4B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":8.4,"file":"BF16","repo":"bartowski/UserRepGen-4B-GGUF","kind":"GGUF","community":false}},"summary":"UserRepGen-4B is a fine-tuned Qwen3.5-4B language model that generates a structured, natural-language user representation from a multi-turn user-assistant conversation. Dense, 4.5B params. Context: 256K tokens (config). License: mit.","est":{"weights":{"Q4":2.7,"Q8":4.5,"FP16":8.4},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"bartowski/UserRepGen-4B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"bartowski/UserRepGen-4B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"bartowski/UserRepGen-4B-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":4.8,"Q8":6.6,"FP16":10.5},"fit":{"nv8":"Q8","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"inclusionAI/Ling-3.0-flash-UBTree","name":"Ling-3.0-flash-UBTree","org":"inclusionAI","date":"2026-10-14","license":"mit","params":481983774,"params_src":"safetensors","active":null,"ctx":131072,"kv_bytes_tok":81920,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"Ling3UBTreeModel","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":5076,"used_storage":964205379,"created":"2026-10-14T10:39:01Z","first_seen":"2026-10-21T23:23:07Z","fresh":false,"url":"https://huggingface.co/inclusionAI/Ling-3.0-flash-UBTree","likes":11,"downloads":191,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.8,"lane":"main","quant_sizes":{"FP16":{"gb":1.0,"file":"official BF16 safetensors","repo":"inclusionAI/Ling-3.0-flash-UBTree","kind":"safetensors","community":false}},"summary":"Text generation model from inclusionAI (HF pipeline tag; the card has no one-line description). Dense, 482M params. Context: 128K tokens (config). License: mit.","est":{"weights":{"Q4":0.3,"Q8":0.5,"FP16":1.0},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.5B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.5B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"inclusionAI/Ling-3.0-flash-UBTree","fmt":"safetensors","community":false}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":2.0,"Q8":2.2,"FP16":2.7},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"amazon/LAR-27B","name":"LAR-27B","org":"amazon","date":"2026-10-14","license":"apache-2.0","params":26895998464,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16"],"pipeline":"","modality":"","summary_what":"","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":4106,"used_storage":53805544683,"created":"2026-10-14T17:31:01Z","first_seen":"2026-10-14T17:31:01Z","fresh":false,"url":"https://huggingface.co/amazon/LAR-27B","likes":6,"downloads":70,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.31,"lane":"main","quant_sizes":{"Q4":{"gb":16.5,"file":"Q4_K_M","repo":"mradermacher/LAR-27B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":28.6,"file":"Q8_0","repo":"mradermacher/LAR-27B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":53.8,"file":"official BF16 safetensors","repo":"amazon/LAR-27B","kind":"safetensors","community":false}},"summary":"Dense, 27B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":16.5,"Q8":28.6,"FP16":53.8},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/LAR-27B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/LAR-27B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"amazon/LAR-27B","fmt":"safetensors","community":false}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":19.6,"Q8":31.7,"FP16":56.9},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"amazon/LAR-9B","name":"LAR-9B","org":"amazon","date":"2026-10-14","license":"apache-2.0","params":8953803264,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":131072,"dtypes":["BF16"],"pipeline":"","modality":"","summary_what":"","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":4098,"used_storage":17921097145,"created":"2026-10-14T17:30:43Z","first_seen":"2026-10-14T17:30:43Z","fresh":false,"url":"https://huggingface.co/amazon/LAR-9B","likes":3,"downloads":33,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.15,"lane":"main","quant_sizes":{"Q4":{"gb":5.6,"file":"Q4_K_M","repo":"mradermacher/LAR-9B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":9.5,"file":"Q8_0","repo":"mradermacher/LAR-9B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":17.9,"file":"F16","repo":"mradermacher/LAR-9B-GGUF","kind":"GGUF","community":false}},"summary":"Dense, 9B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":5.6,"Q8":9.5,"FP16":17.9},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/LAR-9B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/LAR-9B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"mradermacher/LAR-9B-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":7.7,"Q8":11.6,"FP16":20.0},"fit":{"nv8":null,"gpu12":"Q4","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"Q8","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"nvidia/cdf-sim2real-smt-solderlight","name":"cdf-sim2real-smt-solderlight","org":"nvidia","date":"2026-10-14","license":"openmdw-1.1","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":12800,"used_storage":57710671726,"created":"2026-10-14T07:07:34Z","first_seen":"2026-10-20T11:19:23Z","fresh":false,"url":"https://huggingface.co/nvidia/cdf-sim2real-smt-solderlight","likes":0,"downloads":2,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.01,"lane":"main","quant_sizes":{},"summary":"License: openmdw-1.1.","est":null,"license_class":"permissive"},{"type":"model","id":"black-forest-labs/flux-3-image-klein","name":"flux-3-image-klein","org":"black-forest-labs","date":"2026-10-13","license":"flux-kommunity-license","params":5548681984,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"image-to-image","modality":"image","summary_what":"FLUX 3 Image Klein is a 5.6B open weights model for text-to-image generation and image editing.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":14294,"used_storage":43319416391,"created":"2026-10-13T10:58:40Z","first_seen":"2026-10-13T10:58:40Z","fresh":false,"url":"https://huggingface.co/black-forest-labs/flux-3-image-klein","likes":671,"downloads":112091,"finetune":false,"gated":true,"quants":[{"id":"unsloth/flux-3-image-klein-FP8","kind":"Unsloth"},{"id":"unsloth/flux-3-image-klein-GGUF","kind":"GGUF"}],"x_mentions":[{"url":"https://x.com/wwardenn/status/2112268036543578578","user":"wwardenn","likes":22},{"url":"https://x.com/saakin_19/status/2111989646389395507","user":"saakin_19","likes":0},{"url":"https://x.com/conradlotz/status/2111515348503535762","user":"conradlotz","likes":0}],"x_likes":22,"x_posts":3,"signal":"trusted org","velocity":437.95,"lane":"main","quant_sizes":{"Q4":{"gb":3.7,"file":"Q4_K_M","repo":"unsloth/flux-3-image-klein-GGUF","kind":"GGUF","community":false},"Q8":{"gb":6.1,"file":"Q8_0","repo":"unsloth/flux-3-image-klein-GGUF","kind":"GGUF","community":false},"FP16":{"gb":11.1,"file":"BF16","repo":"unsloth/flux-3-image-klein-GGUF","kind":"GGUF","community":false}},"summary":"FLUX 3 Image Klein is a 5.6B open weights model for text-to-image generation and image editing. 5.5B params. License: flux-kommunity-license.","est":{"weights":{"Q4":3.7,"Q8":6.1,"FP16":11.1},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"unsloth/flux-3-image-klein-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"unsloth/flux-3-image-klein-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"unsloth/flux-3-image-klein-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":5.2,"Q8":7.6,"FP16":12.6},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"restricted"},{"type":"model","id":"black-forest-labs/flux-3-image-klein-base","name":"flux-3-image-klein-base","org":"black-forest-labs","date":"2026-10-13","license":"flux-kommunity-license","params":5538458368,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"image-to-image","modality":"image","summary_what":"FLUX 3 Image Klein is a 5.6B open weights model for text-to-image generation and image editing.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":14477,"used_storage":47360928583,"created":"2026-10-13T10:19:46Z","first_seen":"2026-10-13T10:19:46Z","fresh":false,"url":"https://huggingface.co/black-forest-labs/flux-3-image-klein-base","likes":176,"downloads":108779,"finetune":false,"gated":true,"quants":[{"id":"unsloth/flux-3-image-klein-base-FP8","kind":"Unsloth"},{"id":"unsloth/flux-3-image-klein-base-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":393.25,"lane":"main","quant_sizes":{"Q4":{"gb":3.6,"file":"Q4_K_M","repo":"unsloth/flux-3-image-klein-base-GGUF","kind":"GGUF","community":false},"Q8":{"gb":6.1,"file":"Q8_0","repo":"unsloth/flux-3-image-klein-base-GGUF","kind":"GGUF","community":false},"FP16":{"gb":11.1,"file":"BF16","repo":"unsloth/flux-3-image-klein-base-GGUF","kind":"GGUF","community":false}},"summary":"FLUX 3 Image Klein is a 5.6B open weights model for text-to-image generation and image editing. 5.5B params. License: flux-kommunity-license.","est":{"weights":{"Q4":3.6,"Q8":6.1,"FP16":11.1},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"unsloth/flux-3-image-klein-base-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"unsloth/flux-3-image-klein-base-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"unsloth/flux-3-image-klein-base-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":5.1,"Q8":7.6,"FP16":12.6},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"restricted"},{"type":"model","id":"inclusionAI/Ling-3.1-flash","name":"Ling-3.1-flash","org":"inclusionAI","date":"2026-10-13","license":"mit","params":572120680768,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":82944,"dtypes":["BF16","F8_E4M3","F32"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"BailingMoeV3ForCausalLM","type":"MoE","experts":512,"experts_active":8},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":6587,"used_storage":573813820876,"created":"2026-10-13T11:16:41Z","first_seen":"2026-10-13T11:16:41Z","fresh":false,"url":"https://huggingface.co/inclusionAI/Ling-3.1-flash","likes":150,"downloads":4037,"finetune":false,"gated":false,"quants":[{"id":"bartowski/Ling-3.1-flash-GGUF","kind":"GGUF"}],"x_mentions":[{"url":"https://x.com/AntLingAGI/status/2113226533917864148","user":"AntLingAGI","likes":39},{"url":"https://x.com/TeksEdge/status/2113319166908191189","user":"TeksEdge","likes":7},{"url":"https://x.com/runcardsdev/status/2111549969828114702","user":"runcardsdev","likes":0}],"x_likes":48,"x_posts":5,"signal":"trusted org","velocity":65.16,"lane":"main","quant_sizes":{"Q4":{"gb":358.0,"file":"Q4_K_M","repo":"bartowski/Ling-3.1-flash-GGUF","kind":"GGUF","community":false},"Q8":{"gb":608.4,"file":"Q8_0","repo":"bartowski/Ling-3.1-flash-GGUF","kind":"GGUF","community":false}},"summary":"Text generation model from inclusionAI (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 572B total (512 experts, 8 per token). Context: 256K tokens (config). License: mit.","est":{"weights":{"Q4":358.0,"Q8":608.4,"FP16":1144.2},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"bartowski/Ling-3.1-flash-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"bartowski/Ling-3.1-flash-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 572.1B params"}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":359.7,"Q8":610.1,"FP16":1145.9},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":null,"amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":null,"mac512":"Q4","cpu32":null,"cpu64":null,"cpu128":null,"cpu256":null},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":null,"amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":null,"mac512":"real","cpu32":null,"cpu64":null,"cpu128":null,"cpu256":null},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"reflection/Beam-1.0-501B-A23B-DFlash2","name":"Beam-1.0-501B-A23B-DFlash2","org":"reflection","date":"2026-10-13","license":"apache-2.0","params":1927484672,"params_src":"safetensors","active":null,"ctx":1048576,"kv_bytes_tok":20480,"dtypes":["BF16"],"pipeline":"","modality":"","summary_what":"DFlash2 draft model for Beam.","arch":{"class":"DFlash2BeamDraftModel","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19883,"used_storage":3854978296,"created":"2026-10-13T17:17:43Z","first_seen":"2026-10-13T17:17:43Z","fresh":false,"url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B-DFlash2","likes":34,"downloads":1076,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":4.11,"lane":"main","quant_sizes":{"FP16":{"gb":3.9,"file":"official BF16 safetensors","repo":"reflection/Beam-1.0-501B-A23B-DFlash2","kind":"safetensors","community":false}},"summary":"DFlash2 draft model for Beam. Dense, 1.9B params. Context: 1M tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/beam","est":{"weights":{"Q4":1.2,"Q8":2.0,"FP16":3.9},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 1.9B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 1.9B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"reflection/Beam-1.0-501B-A23B-DFlash2","fmt":"safetensors","community":false}},"kv_gb":0.2,"kv_src":"config","need":{"Q4":2.4,"Q8":3.2,"FP16":5.1},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"nvidia/Kimodo-SOMA-RP-ArdyRep-v1.2","name":"Kimodo-SOMA-RP-ArdyRep-v1.2","org":"nvidia","date":"2026-10-13","license":"nvidia-open-model-license","params":283266414,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"","modality":"","summary_what":"The Kimodo model family includes models trained on different skeletons and datasets: Kimodo-SOMA-RP Trained on the 30-joint SOMA skeleton with the proprietary Bones Rigplay dataset.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":13033,"used_storage":1133122928,"created":"2026-10-13T18:04:43Z","first_seen":"2026-10-13T18:04:43Z","fresh":false,"url":"https://huggingface.co/nvidia/Kimodo-SOMA-RP-ArdyRep-v1.2","likes":13,"downloads":148,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.6,"lane":"main","quant_sizes":{},"summary":"The Kimodo model family includes models trained on different skeletons and datasets: Kimodo-SOMA-RP Trained on the 30-joint SOMA skeleton with the proprietary Bones Rigplay dataset. 283M params. License: nvidia-open-model-license.","est":{"weights":{"Q4":0.2,"Q8":0.3,"FP16":0.6},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.3B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.3B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.3B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.7,"Q8":1.8,"FP16":2.1},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"nvidia/Kimodo-SOMA-RP-v1.2","name":"Kimodo-SOMA-RP-v1.2","org":"nvidia","date":"2026-10-13","license":"nvidia-open-model-license","params":283281777,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"","modality":"","summary_what":"The Kimodo model family includes models trained on different skeletons and datasets: Kimodo-SOMA-RP Trained on the 30-joint SOMA skeleton with the proprietary Bones Rigplay dataset.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":12757,"used_storage":1133184460,"created":"2026-10-13T18:03:31Z","first_seen":"2026-10-13T18:03:31Z","fresh":false,"url":"https://huggingface.co/nvidia/Kimodo-SOMA-RP-v1.2","likes":10,"downloads":2331,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":8.69,"lane":"main","quant_sizes":{},"summary":"The Kimodo model family includes models trained on different skeletons and datasets: Kimodo-SOMA-RP Trained on the 30-joint SOMA skeleton with the proprietary Bones Rigplay dataset. 283M params. License: nvidia-open-model-license.","est":{"weights":{"Q4":0.2,"Q8":0.3,"FP16":0.6},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.3B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.3B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.3B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.7,"Q8":1.8,"FP16":2.1},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"nvidia/ARDY-SOMA-RP-30FPS-KimodoRep-Horizon60","name":"ARDY-SOMA-RP-30FPS-KimodoRep-Horizon60","org":"nvidia","date":"2026-10-13","license":"nvidia-open-model-agreement","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":13409,"used_storage":769390360,"created":"2026-10-13T19:17:25Z","first_seen":"2026-10-13T19:17:25Z","fresh":false,"url":"https://huggingface.co/nvidia/ARDY-SOMA-RP-30FPS-KimodoRep-Horizon60","likes":2,"downloads":85,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.32,"lane":"main","quant_sizes":{},"summary":"License: nvidia-open-model-agreement.","est":null,"license_class":"restricted"},{"type":"model","id":"nvidia/Cosmos-AnomalyGen-Glass-Nano","name":"Cosmos-AnomalyGen-Glass-Nano","org":"nvidia","date":"2026-10-13","license":"openmdw-1.1","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"","arch":{},"claims":[],"has_safetensors":false,"has_weights":true,"card_len":11084,"used_storage":46110217,"created":"2026-10-13T04:26:59Z","first_seen":"2026-10-20T11:19:23Z","fresh":false,"url":"https://huggingface.co/nvidia/Cosmos-AnomalyGen-Glass-Nano","likes":0,"downloads":0,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.0,"lane":"main","quant_sizes":{},"summary":"License: openmdw-1.1.","est":null,"license_class":"permissive"},{"type":"model","id":"nvidia/Cosmos-AnomalyGen-Metal-Nano","name":"Cosmos-AnomalyGen-Metal-Nano","org":"nvidia","date":"2026-10-13","license":"openmdw-1.1","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"","arch":{},"claims":[],"has_safetensors":false,"has_weights":true,"card_len":11461,"used_storage":76781321,"created":"2026-10-13T04:26:53Z","first_seen":"2026-10-20T11:19:23Z","fresh":false,"url":"https://huggingface.co/nvidia/Cosmos-AnomalyGen-Metal-Nano","likes":0,"downloads":0,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.0,"lane":"main","quant_sizes":{},"summary":"License: openmdw-1.1.","est":null,"license_class":"permissive"},{"type":"model","id":"nvidia/Cosmos-AnomalyGen-PCB-Nano","name":"Cosmos-AnomalyGen-PCB-Nano","org":"nvidia","date":"2026-10-13","license":"openmdw-1.1","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"","arch":{},"claims":[],"has_safetensors":false,"has_weights":true,"card_len":11853,"used_storage":46110281,"created":"2026-10-13T03:42:40Z","first_seen":"2026-10-20T11:19:23Z","fresh":false,"url":"https://huggingface.co/nvidia/Cosmos-AnomalyGen-PCB-Nano","likes":0,"downloads":0,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.0,"lane":"main","quant_sizes":{},"summary":"License: openmdw-1.1.","est":null,"license_class":"permissive"},{"type":"model","id":"LiquidAI/LFM2.5-24B-A2B","name":"LFM2.5-24B-A2B","org":"LiquidAI","date":"2026-10-12","license":"lfm1.0","params":23971585280,"params_src":"safetensors","active":2000000000.0,"ctx":131072,"kv_bytes_tok":81920,"dtypes":["F32","BF16"],"pipeline":"text-generation","modality":"text","summary_what":"LFM2.5-24B-A2B is a 24B-parameter reasoning model with only 2B active parameters, built for fast, private agents on-device.","arch":{"class":"Lfm2MoeForCausalLM","type":"MoE","experts":64,"experts_active":4,"active":2000000000.0,"active_src":"name"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19708,"used_storage":47965588445,"created":"2026-10-12T05:35:46Z","first_seen":"2026-10-12T05:35:46Z","fresh":false,"url":"https://huggingface.co/LiquidAI/LFM2.5-24B-A2B","likes":578,"downloads":19800,"finetune":true,"gated":false,"quants":[{"id":"LiquidAI/LFM2.5-24B-A2B-FP8","kind":"quant"},{"id":"LiquidAI/LFM2.5-24B-A2B-GGUF","kind":"GGUF"},{"id":"LiquidAI/LFM2.5-24B-A2B-MLX-4bit","kind":"MLX"},{"id":"LiquidAI/LFM2.5-24B-A2B-MLX-5bit","kind":"MLX"},{"id":"LiquidAI/LFM2.5-24B-A2B-MLX-6bit","kind":"MLX"},{"id":"LiquidAI/LFM2.5-24B-A2B-MLX-8bit","kind":"MLX"},{"id":"LiquidAI/LFM2.5-24B-A2B-MLX-bf16","kind":"MLX"},{"id":"LiquidAI/LFM2.5-24B-A2B-ONNX","kind":"quant"},{"id":"bartowski/LFM2.5-24B-A2B-GGUF","kind":"GGUF"},{"id":"ggml-org/LFM2.5-24B-A2B-GGUF","kind":"GGUF"},{"id":"mlx-community/LFM2.5-24B-A2B-OptiQ-4bit","kind":"MLX"},{"id":"unsloth/LFM2.5-24B-A2B","kind":"Unsloth"}],"x_mentions":[{"url":"https://x.com/Oluwaphilemon1/status/2110917767045972095","user":"Oluwaphilemon1","likes":4},{"url":"https://x.com/dextune/status/2110178099296121074","user":"dextune","likes":4},{"url":"https://x.com/jvr0x/status/2112817955138326575","user":"jvr0x","likes":2}],"x_likes":19,"x_posts":5,"signal":"trusted org","velocity":116.64,"lane":"main","quant_sizes":{"Q4":{"gb":14.5,"file":"Q4_K_M","repo":"LiquidAI/LFM2.5-24B-A2B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":25.5,"file":"Q8_0","repo":"LiquidAI/LFM2.5-24B-A2B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":48.0,"file":"BF16","repo":"LiquidAI/LFM2.5-24B-A2B-GGUF","kind":"GGUF","community":false}},"summary":"LFM2.5-24B-A2B is a 24B-parameter reasoning model with only 2B active parameters, built for fast, private agents on-device. Mixture-of-experts, 24B total, 2B active (64 experts, 4 per token). Context: 128K tokens (config). License: lfm1.0.","ollama":"https://ollama.com/library/lfm2.5","est":{"weights":{"Q4":14.5,"Q8":25.5,"FP16":48.0},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"LiquidAI/LFM2.5-24B-A2B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"LiquidAI/LFM2.5-24B-A2B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"LiquidAI/LFM2.5-24B-A2B-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":16.2,"Q8":27.2,"FP16":49.7},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q8","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":"Q8","mac16":null,"mac24":"Q4","mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":"real","mac16":null,"mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"restricted"},{"type":"model","id":"nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16","name":"NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16","org":"nvidia","date":"2026-10-12","license":"openmdw-1.1","params":124453395976,"params_src":"safetensors","active":12000000000.0,"ctx":1048576,"kv_bytes_tok":null,"dtypes":["BF16","F32","I64"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 is the full-precision release of NVIDIA Nemotron 3.5 Super VL, a vision-language model for reasoning, coding, tool use, document understanding, and video understanding.","arch":{"class":"NemotronH_Omni_Reasoning_V3","type":"MoE","experts":512,"experts_active":22,"active":12000000000.0,"active_src":"name"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19656,"used_storage":248932364645,"created":"2026-10-12T15:32:06Z","first_seen":"2026-10-12T15:32:06Z","fresh":false,"url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16","likes":99,"downloads":31694,"finetune":false,"gated":false,"quants":[{"id":"ggml-org/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-GGUF","kind":"GGUF"},{"id":"nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4","kind":"quant"},{"id":"nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4-LiLiCorr","kind":"quant"},{"id":"unsloth/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B","kind":"Unsloth"},{"id":"unsloth/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-GGUF","kind":"GGUF"},{"id":"unsloth/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-NVFP4","kind":"Unsloth"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":107.46,"lane":"main","quant_sizes":{"Q4":{"gb":86.9,"file":"UD-Q4_K_XL","repo":"unsloth/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":131.6,"file":"Q8_0","repo":"unsloth/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":249.6,"file":"BF16","repo":"ggml-org/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-GGUF","kind":"GGUF","community":false}},"summary":"NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16 is the full-precision release of NVIDIA Nemotron 3.5 Super VL, a vision-language model for reasoning, coding, tool use, document understanding, and video understanding. Mixture-of-experts, 124B total, 12B active (512 experts, 22 per token). Context: 1M tokens (config). License: openmdw-1.1.","ollama":"https://ollama.com/library/nemotron-3.5-super","est":{"weights":{"Q4":86.9,"Q8":131.6,"FP16":249.6},"weights_src":{"Q4":{"kind":"real","file":"UD-Q4_K_XL","repo":"unsloth/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"unsloth/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"ggml-org/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-GGUF","fmt":"GGUF","community":false}},"kv_gb":10.6,"kv_src":"rough","need":{"Q4":98.5,"Q8":143.2,"FP16":261.2},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":"Q4","spark2":"Q8","amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":"Q8","mac256":"Q8","mac512":"FP16","cpu32":null,"cpu64":null,"cpu128":"Q4","cpu256":"Q8"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":"real","mac256":"real","mac512":"real","cpu32":null,"cpu64":null,"cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-Base-BF16","name":"NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-Base-BF16","org":"nvidia","date":"2026-10-12","license":"openmdw-1.1","params":124453395974,"params_src":"safetensors","active":12000000000.0,"ctx":262144,"kv_bytes_tok":null,"dtypes":["BF16","F32"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-Base-BF16 is the base vision-language checkpoint of the Nemotron 3.5 Super family.","arch":{"class":"NemotronH_Omni_Reasoning_V3","type":"MoE","experts":512,"experts_active":22,"active":12000000000.0,"active_src":"name"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19681,"used_storage":248932364541,"created":"2026-10-12T15:31:49Z","first_seen":"2026-10-12T15:31:49Z","fresh":false,"url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-Base-BF16","likes":54,"downloads":1760,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":6.13,"lane":"main","quant_sizes":{"FP16":{"gb":248.9,"file":"official BF16 safetensors","repo":"nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-Base-BF16","kind":"safetensors","community":false}},"summary":"NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-Base-BF16 is the base vision-language checkpoint of the Nemotron 3.5 Super family. Mixture-of-experts, 124B total, 12B active (512 experts, 22 per token). Context: 256K tokens (config). License: openmdw-1.1.","ollama":"https://ollama.com/library/nemotron-3.5-super","est":{"weights":{"Q4":74.7,"Q8":132.2,"FP16":248.9},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 124.5B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 124.5B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"nvidia/NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-Base-BF16","fmt":"safetensors","community":false}},"kv_gb":10.6,"kv_src":"rough","need":{"Q4":86.3,"Q8":143.8,"FP16":260.5},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":"Q4","spark1":"Q4","spark2":"Q8","amd16":null,"amd24":null,"strix":"Q4","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":"Q4","mac192":"Q8","mac256":"Q8","mac512":"FP16","cpu32":null,"cpu64":null,"cpu128":"Q4","cpu256":"Q8"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":null,"amd24":null,"strix":"estimate","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"real","cpu32":null,"cpu64":null,"cpu128":"estimate","cpu256":"estimate"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"theforecastingcompany/t0-gamma","name":"t0-gamma","org":"theforecastingcompany","date":"2026-10-12","license":"apache-2.0","params":255614293,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"time-series-forecasting","modality":"other","summary_what":"","arch":{"type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":15139,"used_storage":1022492908,"created":"2026-10-12T14:37:40Z","first_seen":"2026-10-12T14:37:40Z","fresh":false,"url":"https://huggingface.co/theforecastingcompany/t0-gamma","likes":24,"downloads":21118,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"21118 downloads","velocity":71.24,"lane":"main","quant_sizes":{},"summary":"Time series forecasting model from theforecastingcompany (HF pipeline tag; the card has no one-line description). Dense, 256M params. License: apache-2.0.","est":{"weights":{"Q4":0.2,"Q8":0.3,"FP16":0.5},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.3B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.3B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.3B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.7,"Q8":1.8,"FP16":2.0},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"openbmb/Trident-Qwen3VL-2B","name":"Trident-Qwen3VL-2B","org":"openbmb","date":"2026-10-12","license":"mit","params":2127532032,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":114688,"dtypes":["BF16"],"pipeline":"feature-extraction","modality":"embedding","summary_what":"Trident-Qwen3VL-2B is a multimodal embedding model for mixed-modality retrieval.","arch":{"class":"Qwen3VLForEmbedding","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":5350,"used_storage":4266563233,"created":"2026-10-12T10:25:38Z","first_seen":"2026-10-12T10:25:38Z","fresh":false,"url":"https://huggingface.co/openbmb/Trident-Qwen3VL-2B","likes":10,"downloads":589,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":1.99,"lane":"main","quant_sizes":{"FP16":{"gb":4.3,"file":"official BF16 safetensors","repo":"openbmb/Trident-Qwen3VL-2B","kind":"safetensors","community":false}},"summary":"Trident-Qwen3VL-2B is a multimodal embedding model for mixed-modality retrieval. Dense, 2.1B params. Context: 256K tokens (config). License: mit.","ollama":"https://ollama.com/library/qwen3-vl","est":{"weights":{"Q4":1.3,"Q8":2.3,"FP16":4.3},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 2.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 2.1B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"openbmb/Trident-Qwen3VL-2B","fmt":"safetensors","community":false}},"kv_gb":0.9,"kv_src":"config","need":{"Q4":3.2,"Q8":4.2,"FP16":6.2},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"allenai/HiRO-ACEv2","name":"HiRO-ACEv2","org":"allenai","date":"2026-10-12","license":"apache-2.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"HiRO (High Resolution Output) is a diffusion model which generates downscaled fields at 3 km resolution from 100 km resolution inputs.","arch":{},"claims":[],"has_safetensors":false,"has_weights":true,"card_len":5285,"used_storage":10908523756,"created":"2026-10-12T16:43:07Z","first_seen":"2026-10-12T16:43:07Z","fresh":false,"url":"https://huggingface.co/allenai/HiRO-ACEv2","likes":5,"downloads":0,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.02,"lane":"main","quant_sizes":{},"summary":"HiRO (High Resolution Output) is a diffusion model which generates downscaled fields at 3 km resolution from 100 km resolution inputs. License: apache-2.0.","est":null,"license_class":"permissive"},{"type":"model","id":"OpenMOSS-Team/MOSS-Image-SFT","name":"MOSS-Image-SFT","org":"OpenMOSS-Team","date":"2026-10-11","license":"apache-2.0","params":5082308624,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16","F32"],"pipeline":"text-to-image","modality":"image","summary_what":"MOSS-Image-SFT is a supervised fine-tuned text-to-image model in the MOSS-Image family from OpenMOSS, derived from MOSS-Image-Base.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":8954,"used_storage":27228981820,"created":"2026-10-11T07:01:17Z","first_seen":"2026-10-11T07:01:17Z","fresh":false,"url":"https://huggingface.co/OpenMOSS-Team/MOSS-Image-SFT","likes":687,"downloads":9717,"finetune":true,"gated":false,"quants":[{"id":"unsloth/MOSS-Image-SFT-FP8","kind":"Unsloth"},{"id":"unsloth/MOSS-Image-SFT-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"687 likes, 9717 downloads","velocity":31.68,"lane":"main","quant_sizes":{"Q4":{"gb":3.2,"file":"Q4_K_M","repo":"unsloth/MOSS-Image-SFT-GGUF","kind":"GGUF","community":false},"Q8":{"gb":5.5,"file":"Q8_0","repo":"unsloth/MOSS-Image-SFT-GGUF","kind":"GGUF","community":false},"FP16":{"gb":10.2,"file":"BF16","repo":"unsloth/MOSS-Image-SFT-GGUF","kind":"GGUF","community":false}},"summary":"MOSS-Image-SFT is a supervised fine-tuned text-to-image model in the MOSS-Image family from OpenMOSS, derived from MOSS-Image-Base. 5.1B params. License: apache-2.0.","est":{"weights":{"Q4":3.2,"Q8":5.5,"FP16":10.2},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"unsloth/MOSS-Image-SFT-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"unsloth/MOSS-Image-SFT-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"unsloth/MOSS-Image-SFT-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":4.7,"Q8":7.0,"FP16":11.7},"fit":{"nv8":"Q8","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"OpenMOSS-Team/MOSS-Image-Base","name":"MOSS-Image-Base","org":"OpenMOSS-Team","date":"2026-10-11","license":"apache-2.0","params":5082308624,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16","F32"],"pipeline":"text-to-image","modality":"image","summary_what":"MOSS-Image-Base is the base text-to-image model in the MOSS-Image family from OpenMOSS.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":8803,"used_storage":27228815653,"created":"2026-10-11T06:49:17Z","first_seen":"2026-10-11T06:49:17Z","fresh":false,"url":"https://huggingface.co/OpenMOSS-Team/MOSS-Image-Base","likes":533,"downloads":6943,"finetune":false,"gated":false,"quants":[{"id":"unsloth/MOSS-Image-Base-FP8","kind":"Unsloth"},{"id":"unsloth/MOSS-Image-Base-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"533 likes, 6943 downloads","velocity":22.75,"lane":"main","quant_sizes":{"Q4":{"gb":3.2,"file":"Q4_K_M","repo":"unsloth/MOSS-Image-Base-GGUF","kind":"GGUF","community":false},"Q8":{"gb":5.5,"file":"Q8_0","repo":"unsloth/MOSS-Image-Base-GGUF","kind":"GGUF","community":false},"FP16":{"gb":10.2,"file":"BF16","repo":"unsloth/MOSS-Image-Base-GGUF","kind":"GGUF","community":false}},"summary":"MOSS-Image-Base is the base text-to-image model in the MOSS-Image family from OpenMOSS. 5.1B params. License: apache-2.0.","est":{"weights":{"Q4":3.2,"Q8":5.5,"FP16":10.2},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"unsloth/MOSS-Image-Base-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"unsloth/MOSS-Image-Base-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"unsloth/MOSS-Image-Base-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":4.7,"Q8":7.0,"FP16":11.7},"fit":{"nv8":"Q8","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"datalab-to/indra-ocr","name":"indra-ocr","org":"datalab-to","date":"2026-10-10","license":"openrail","params":1250566,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"image-to-text","modality":"vision","summary_what":"Indra OCR is a 1M-parameter OCR model that runs on CPU.","arch":{},"claims":["When you need the highest accuracy, on handwriting, math, complex tables and degraded scans, our managed platform runs Chandra, our state of the art OCR model, with zero data retention by default, SOC 2 Type 2, and custom BAAs."],"has_safetensors":true,"has_weights":true,"card_len":19378,"used_storage":127714196,"created":"2026-10-10T14:12:33Z","first_seen":"2026-10-10T14:12:33Z","fresh":false,"url":"https://huggingface.co/datalab-to/indra-ocr","likes":401,"downloads":4436,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"401 likes","velocity":14.01,"lane":"main","quant_sizes":{},"summary":"Indra OCR is a 1M-parameter OCR model that runs on CPU. 1M params. License: openrail. Vendor-reported: \"When you need the highest accuracy, on handwriting, math, complex tables and degraded scans, our managed platform runs Chandra, our state of the art OCR model, with zero data retention by default, SOC 2 Type 2, and custom BAAs.\"","est":{"weights":{"Q4":0.0,"Q8":0.0,"FP16":0.0},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.0B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.0B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.0B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.5,"Q8":1.5,"FP16":1.5},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"vllm-sr/d3","name":"d3","org":"vllm-sr","date":"2026-10-10","license":"apache-2.0","params":26085330160,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16"],"pipeline":"zero-shot-classification","modality":"other","summary_what":"","arch":{"class":"Qwen3_5Model","type":"dense"},"claims":["d3 ranks first of the 149 entries on the official text board and is on its Pareto frontier: no entry with as many parameters or fewer scores higher."],"has_safetensors":true,"has_weights":true,"card_len":19752,"used_storage":208723144639,"created":"2026-10-10T09:38:30Z","first_seen":"2026-10-10T09:38:30Z","fresh":false,"url":"https://huggingface.co/vllm-sr/d3","likes":268,"downloads":6949,"finetune":true,"gated":false,"quants":[{"id":"ggml-org/vllm-d3-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"268 likes, 6949 downloads","velocity":20.63,"lane":"main","quant_sizes":{"Q4":{"gb":19.4,"file":"Q4_K_M","repo":"ggml-org/vllm-d3-GGUF","kind":"GGUF","community":false},"Q8":{"gb":28.6,"file":"Q8_0","repo":"ggml-org/vllm-d3-GGUF","kind":"GGUF","community":false},"FP16":{"gb":53.8,"file":"BF16","repo":"ggml-org/vllm-d3-GGUF","kind":"GGUF","community":false}},"summary":"Zero shot classification model from vllm-sr (HF pipeline tag; the card has no one-line description). Dense, 26B params. Context: 256K tokens (config). License: apache-2.0. Vendor-reported: \"d3 ranks first of the 149 entries on the official text board and is on its Pareto frontier: no entry with as many parameters or fewer scores higher.\"","ollama":"https://ollama.com/library/d3","est":{"weights":{"Q4":19.4,"Q8":28.6,"FP16":53.8},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"ggml-org/vllm-d3-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"ggml-org/vllm-d3-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"ggml-org/vllm-d3-GGUF","fmt":"GGUF","community":false}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":22.5,"Q8":31.7,"FP16":56.9},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"dmis-lab/UltraBERT-380M","name":"UltraBERT-380M","org":"dmis-lab","date":"2026-10-10","license":"apache-2.0","params":429966336,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":98304,"dtypes":["F32"],"pipeline":"fill-mask","modality":"other","summary_what":"UltraBERT is a family of English bidirectional encoders with a native 32K context that reach on both classification and retrieval, two task families on which previous encoders had to choose a side.","arch":{"class":"UltrabertForMaskedLM","type":"dense"},"claims":["UltraBERT is a family of English bidirectional encoders with a native 32K context that reach state of the art on both classification and retrieval, two task families on which previous encoders had to choose a side."],"has_safetensors":true,"has_weights":true,"card_len":11708,"used_storage":1931342314,"created":"2026-10-10T10:53:18Z","first_seen":"2026-10-10T10:53:18Z","fresh":false,"url":"https://huggingface.co/dmis-lab/UltraBERT-380M","likes":72,"downloads":9825,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"72 likes, 9825 downloads","velocity":28.4,"lane":"main","quant_sizes":{},"summary":"UltraBERT is a family of English bidirectional encoders with a native 32K context that reach on both classification and retrieval, two task families on which previous encoders had to choose a side. Dense, 430M params. License: apache-2.0. Vendor-reported: \"UltraBERT is a family of English bidirectional encoders with a native 32K context that reach state of the art on both classification and retrieval, two task families on which previous encoders had to choose a side.\"","est":{"weights":{"Q4":0.3,"Q8":0.5,"FP16":0.9},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.4B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.4B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.4B params"}},"kv_gb":0.8,"kv_src":"config","need":{"Q4":2.1,"Q8":2.3,"FP16":2.7},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"dmis-lab/UltraBERT-150M","name":"UltraBERT-150M","org":"dmis-lab","date":"2026-10-10","license":"apache-2.0","params":189960960,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":73728,"dtypes":["F32"],"pipeline":"fill-mask","modality":"other","summary_what":"UltraBERT is a family of English bidirectional encoders with a native 32K context that reach on both classification and retrieval, two task families on which previous encoders had to choose a side.","arch":{"class":"UltrabertForMaskedLM","type":"dense"},"claims":["UltraBERT is a family of English bidirectional encoders with a native 32K context that reach state of the art on both classification and retrieval, two task families on which previous encoders had to choose a side."],"has_safetensors":true,"has_weights":true,"card_len":11708,"used_storage":922495901,"created":"2026-10-10T10:52:59Z","first_seen":"2026-10-10T10:52:59Z","fresh":false,"url":"https://huggingface.co/dmis-lab/UltraBERT-150M","likes":59,"downloads":7809,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"59 likes, 7809 downloads","velocity":22.58,"lane":"main","quant_sizes":{},"summary":"UltraBERT is a family of English bidirectional encoders with a native 32K context that reach on both classification and retrieval, two task families on which previous encoders had to choose a side. Dense, 190M params. License: apache-2.0. Vendor-reported: \"UltraBERT is a family of English bidirectional encoders with a native 32K context that reach state of the art on both classification and retrieval, two task families on which previous encoders had to choose a side.\"","est":{"weights":{"Q4":0.1,"Q8":0.2,"FP16":0.4},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.2B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.2B params"}},"kv_gb":0.6,"kv_src":"config","need":{"Q4":1.7,"Q8":1.8,"FP16":2.0},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"nvidia/cosmos3-nano-libero10-sdg-export-S5_2","name":"cosmos3-nano-libero10-sdg-export-S5_2","org":"nvidia","date":"2026-10-10","license":"nvidia-open-model-license","params":15173136576,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"robotics","modality":"robotics","summary_what":"","arch":{"class":"Cosmos3OmniModel"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":1372,"used_storage":31499212848,"created":"2026-10-10T18:54:08Z","first_seen":"2026-10-18T23:26:24Z","fresh":false,"url":"https://huggingface.co/nvidia/cosmos3-nano-libero10-sdg-export-S5_2","likes":0,"downloads":92,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.27,"lane":"main","quant_sizes":{"FP16":{"gb":31.5,"file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-S5_2","kind":"safetensors","community":false}},"summary":"Robotics model from nvidia (HF pipeline tag; the card has no one-line description). 15B params. License: nvidia-open-model-license.","est":{"weights":{"Q4":9.1,"Q8":16.1,"FP16":31.5},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 15.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 15.2B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-S5_2","fmt":"safetensors","community":false}},"kv_gb":1.3,"kv_src":"rough","need":{"Q4":11.4,"Q8":18.4,"FP16":33.8},"fit":{"nv8":null,"gpu12":null,"gpu16":"Q4","nv24":"Q8","nv32":"Q8","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q4","amd24":"Q8","strix":"FP16","gpu24":"Q8","gpu32":"Q8","mac16":"Q4","mac24":"Q4","mac36":"Q8","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q8","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"estimate","amd24":"estimate","strix":"real","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"nvidia/cosmos3-nano-libero10-sdg-export-S5_all","name":"cosmos3-nano-libero10-sdg-export-S5_all","org":"nvidia","date":"2026-10-10","license":"nvidia-open-model-license","params":15173136576,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"robotics","modality":"robotics","summary_what":"","arch":{"class":"Cosmos3OmniModel"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":1350,"used_storage":31499212848,"created":"2026-10-10T18:53:19Z","first_seen":"2026-10-18T23:26:24Z","fresh":false,"url":"https://huggingface.co/nvidia/cosmos3-nano-libero10-sdg-export-S5_all","likes":0,"downloads":96,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.28,"lane":"main","quant_sizes":{"FP16":{"gb":31.5,"file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-S5_all","kind":"safetensors","community":false}},"summary":"Robotics model from nvidia (HF pipeline tag; the card has no one-line description). 15B params. License: nvidia-open-model-license.","est":{"weights":{"Q4":9.1,"Q8":16.1,"FP16":31.5},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 15.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 15.2B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-S5_all","fmt":"safetensors","community":false}},"kv_gb":1.3,"kv_src":"rough","need":{"Q4":11.4,"Q8":18.4,"FP16":33.8},"fit":{"nv8":null,"gpu12":null,"gpu16":"Q4","nv24":"Q8","nv32":"Q8","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q4","amd24":"Q8","strix":"FP16","gpu24":"Q8","gpu32":"Q8","mac16":"Q4","mac24":"Q4","mac36":"Q8","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q8","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"estimate","amd24":"estimate","strix":"real","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"Qwen/Qwen-Image-2.1-Turbo","name":"Qwen-Image-2.1-Turbo","org":"Qwen","date":"2026-10-09","license":"qwen-research","params":7115124736,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"text-to-image","modality":"image","summary_what":"Qwen-Image-2.1-Turbo is an accelerated checkpoint of Qwen-Image-2.1 for text-to-image generation and image editing with 8 denoising steps.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":14815,"used_storage":32494210741,"created":"2026-10-09T04:50:01Z","first_seen":"2026-10-09T04:50:01Z","fresh":false,"url":"https://huggingface.co/Qwen/Qwen-Image-2.1-Turbo","likes":1047,"downloads":32825,"finetune":true,"gated":false,"quants":[{"id":"unsloth/Qwen-Image-2.1-Turbo-FP8","kind":"Unsloth"},{"id":"unsloth/Qwen-Image-2.1-Turbo-GGUF","kind":"GGUF"}],"x_mentions":[{"url":"https://x.com/Oluwaphilemon1/status/2112035566183170252","user":"Oluwaphilemon1","likes":9},{"url":"https://x.com/EaarthNet/status/2111101752804032716","user":"EaarthNet","likes":1},{"url":"https://x.com/philppplik/status/2113236925188362498","user":"philppplik","likes":0}],"x_likes":10,"x_posts":3,"signal":"trusted org","velocity":119.48,"lane":"main","quant_sizes":{"Q4":{"gb":4.2,"file":"Q4_K_M","repo":"unsloth/Qwen-Image-2.1-Turbo-GGUF","kind":"GGUF","community":false},"Q8":{"gb":7.6,"file":"Q8_0","repo":"unsloth/Qwen-Image-2.1-Turbo-GGUF","kind":"GGUF","community":false},"FP16":{"gb":14.2,"file":"F16","repo":"unsloth/Qwen-Image-2.1-Turbo-GGUF","kind":"GGUF","community":false}},"summary":"Qwen-Image-2.1-Turbo is an accelerated checkpoint of Qwen-Image-2.1 for text-to-image generation and image editing with 8 denoising steps. 7.1B params. License: qwen-research.","ollama":"https://ollama.com/library/qwen","est":{"weights":{"Q4":4.2,"Q8":7.6,"FP16":14.2},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"unsloth/Qwen-Image-2.1-Turbo-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"unsloth/Qwen-Image-2.1-Turbo-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"unsloth/Qwen-Image-2.1-Turbo-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.6,"kv_src":"rough","need":{"Q4":5.8,"Q8":9.2,"FP16":15.8},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"restricted"},{"type":"model","id":"Cloudflare/clef-omni","name":"clef-omni","org":"Cloudflare","date":"2026-10-09","license":"apache-2.0","params":35259818545,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"Clef-Omni is a 30B-A3B mixture-of-experts multimodal model that turns a state and a schema of typed questions into decisions.","arch":{"class":"Qwen3OmniMoeForConditionalGeneration"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":10064,"used_storage":70753087464,"created":"2026-10-09T04:10:22Z","first_seen":"2026-10-09T04:10:22Z","fresh":false,"url":"https://huggingface.co/Cloudflare/clef-omni","likes":140,"downloads":6410,"finetune":true,"gated":false,"quants":[{"id":"mlx-community/clef-omni-4bit","kind":"MLX"},{"id":"mlx-community/clef-omni-8bit","kind":"MLX"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"140 likes, 6410 downloads","velocity":17.27,"lane":"main","quant_sizes":{"Q4":{"gb":19.8,"file":"MLX 4-bit","repo":"mlx-community/clef-omni-4bit","kind":"MLX","community":false},"Q8":{"gb":35.0,"file":"MLX 8-bit","repo":"mlx-community/clef-omni-8bit","kind":"MLX","community":false},"FP16":{"gb":70.7,"file":"official BF16 safetensors","repo":"Cloudflare/clef-omni","kind":"safetensors","community":false}},"summary":"Clef-Omni is a 30B-A3B mixture-of-experts multimodal model that turns a state and a schema of typed questions into decisions. 35B params. License: apache-2.0.","ollama":"https://ollama.com/library/clef","est":{"weights":{"Q4":19.8,"Q8":35.0,"FP16":70.7},"weights_src":{"Q4":{"kind":"real","file":"MLX 4-bit","repo":"mlx-community/clef-omni-4bit","fmt":"MLX","community":false},"Q8":{"kind":"real","file":"MLX 8-bit","repo":"mlx-community/clef-omni-8bit","fmt":"MLX","community":false},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"Cloudflare/clef-omni","fmt":"safetensors","community":false}},"kv_gb":3.0,"kv_src":"rough","need":{"Q4":23.8,"Q8":39.0,"FP16":74.7},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":null,"strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q4","mac64":"Q8","mac96":"Q8","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"tencent/Youtu-Parsing-Omni","name":"Youtu-Parsing-Omni","org":"tencent","date":"2026-10-09","license":"youtu-parsing","params":5334472192,"params_src":"safetensors","active":null,"ctx":1048576,"kv_bytes_tok":46080,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"Youtu-Parsing-Omni is a compact (5B) omni-modal parsing model.","arch":{"class":"YoutuVITAForCausalLM","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19710,"used_storage":10672602625,"created":"2026-10-09T06:42:18Z","first_seen":"2026-10-09T06:42:18Z","fresh":false,"url":"https://huggingface.co/tencent/Youtu-Parsing-Omni","likes":115,"downloads":2517,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":6.99,"lane":"main","quant_sizes":{"FP16":{"gb":10.7,"file":"official BF16 safetensors","repo":"tencent/Youtu-Parsing-Omni","kind":"safetensors","community":false}},"summary":"Youtu-Parsing-Omni is a compact (5B) omni-modal parsing model. Dense, 5.3B params. Context: 1M tokens (config). License: youtu-parsing.","est":{"weights":{"Q4":3.2,"Q8":5.7,"FP16":10.7},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 5.3B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 5.3B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"tencent/Youtu-Parsing-Omni","fmt":"safetensors","community":false}},"kv_gb":0.4,"kv_src":"config","need":{"Q4":4.6,"Q8":7.1,"FP16":12.1},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"estimate","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"nvidia/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-BF16","name":"NVIDIA-Nemotron-3.5-Bolt-74B-A9B-BF16","org":"nvidia","date":"2026-10-09","license":"openmdw-1.1","params":74885081096,"params_src":"safetensors","active":9000000000.0,"ctx":262144,"kv_bytes_tok":null,"dtypes":["F32","BF16","I64"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"NVIDIA Nemotron 3.5 Bolt 74B-A9B is a vision-language model (VLM) designed to run everyday agentic tasks locally.","arch":{"class":"NemotronH_Omni_Reasoning_V3","type":"MoE","experts":512,"experts_active":null,"active":9000000000.0,"active_src":"name"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19590,"used_storage":149792249678,"created":"2026-10-09T12:45:14Z","first_seen":"2026-10-09T12:45:14Z","fresh":false,"url":"https://huggingface.co/nvidia/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-BF16","likes":47,"downloads":4474,"finetune":true,"gated":false,"quants":[{"id":"nvidia/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB-GGUF","kind":"GGUF"},{"id":"unsloth/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-GGUF","kind":"GGUF"}],"x_mentions":[{"url":"https://x.com/kaimi_local/status/2111073371517759705","user":"kaimi_local","likes":0}],"x_likes":0,"x_posts":1,"signal":"trusted org","velocity":22.2,"lane":"main","quant_sizes":{"Q4":{"gb":50.2,"file":"UD-Q4_K_XL","repo":"unsloth/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":78.9,"file":"Q8_0","repo":"unsloth/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":148.2,"file":"BF16","repo":"unsloth/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-GGUF","kind":"GGUF","community":false}},"summary":"NVIDIA Nemotron 3.5 Bolt 74B-A9B is a vision-language model (VLM) designed to run everyday agentic tasks locally. Mixture-of-experts, 75B total, 9B active (512 experts). Context: 256K tokens (config). License: openmdw-1.1.","ollama":"https://ollama.com/library/nemotron-3.5-bolt","est":{"weights":{"Q4":50.2,"Q8":78.9,"FP16":148.2},"weights_src":{"Q4":{"kind":"real","file":"UD-Q4_K_XL","repo":"unsloth/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"unsloth/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"unsloth/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-GGUF","fmt":"GGUF","community":false}},"kv_gb":6.4,"kv_src":"rough","need":{"Q4":57.6,"Q8":86.3,"FP16":155.6},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":"Q8","spark1":"Q8","spark2":"FP16","amd16":null,"amd24":null,"strix":"Q8","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":"Q4","mac128":"Q8","mac192":"Q8","mac256":"FP16","mac512":"FP16","cpu32":null,"cpu64":null,"cpu128":"Q8","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":null,"cpu64":null,"cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"LiquidAI/LFM2.5-24B-A2B-Base","name":"LFM2.5-24B-A2B-Base","org":"LiquidAI","date":"2026-10-09","license":"lfm1.0","params":23971585280,"params_src":"safetensors","active":2000000000.0,"ctx":131072,"kv_bytes_tok":81920,"dtypes":["F32","BF16"],"pipeline":"text-generation","modality":"text","summary_what":"LFM2.5 is a new family of hybrid models designed for on-device deployment.","arch":{"class":"Lfm2MoeForCausalLM","type":"MoE","experts":64,"experts_active":4,"active":2000000000.0,"active_src":"name"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":11037,"used_storage":47962049045,"created":"2026-10-09T21:13:55Z","first_seen":"2026-10-09T21:13:55Z","fresh":false,"url":"https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-Base","likes":38,"downloads":1465,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":4.15,"lane":"main","quant_sizes":{"FP16":{"gb":47.9,"file":"official BF16 safetensors","repo":"LiquidAI/LFM2.5-24B-A2B-Base","kind":"safetensors","community":false}},"summary":"LFM2.5 is a new family of hybrid models designed for on-device deployment. Mixture-of-experts, 24B total, 2B active (64 experts, 4 per token). Context: 128K tokens (config). License: lfm1.0.","ollama":"https://ollama.com/library/lfm2.5","est":{"weights":{"Q4":14.4,"Q8":25.5,"FP16":47.9},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 24.0B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 24.0B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"LiquidAI/LFM2.5-24B-A2B-Base","fmt":"safetensors","community":false}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":16.1,"Q8":27.2,"FP16":49.6},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q8","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":"Q8","mac16":null,"mac24":"Q4","mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"estimate","strix":"real","gpu24":null,"gpu32":"estimate","mac16":null,"mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"nvidia/cosmos3-nano-libero10-sdg-export-D5","name":"cosmos3-nano-libero10-sdg-export-D5","org":"nvidia","date":"2026-10-09","license":"nvidia-open-model-license","params":15173136576,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"robotics","modality":"robotics","summary_what":"","arch":{"class":"Cosmos3OmniModel"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":1335,"used_storage":31499212848,"created":"2026-10-09T21:31:12Z","first_seen":"2026-10-18T23:26:24Z","fresh":false,"url":"https://huggingface.co/nvidia/cosmos3-nano-libero10-sdg-export-D5","likes":0,"downloads":99,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.27,"lane":"main","quant_sizes":{"FP16":{"gb":31.5,"file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-D5","kind":"safetensors","community":false}},"summary":"Robotics model from nvidia (HF pipeline tag; the card has no one-line description). 15B params. License: nvidia-open-model-license.","est":{"weights":{"Q4":9.1,"Q8":16.1,"FP16":31.5},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 15.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 15.2B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-D5","fmt":"safetensors","community":false}},"kv_gb":1.3,"kv_src":"rough","need":{"Q4":11.4,"Q8":18.4,"FP16":33.8},"fit":{"nv8":null,"gpu12":null,"gpu16":"Q4","nv24":"Q8","nv32":"Q8","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q4","amd24":"Q8","strix":"FP16","gpu24":"Q8","gpu32":"Q8","mac16":"Q4","mac24":"Q4","mac36":"Q8","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q8","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"estimate","amd24":"estimate","strix":"real","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"cloud0day3/antalia-mini","name":"antalia-mini","org":"cloud0day3","date":"2026-10-08","license":"apache-2.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"text-to-speech","modality":"audio","summary_what":"Antalia-2 Mini (version 1.0.1) is a small Turkish text-to-speech model with one voice.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":18796,"used_storage":110022451,"created":"2026-10-08T10:57:38Z","first_seen":"2026-10-08T10:57:38Z","fresh":false,"url":"https://huggingface.co/cloud0day3/antalia-mini","likes":54,"downloads":5270,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"54 likes, 5270 downloads","velocity":13.43,"lane":"main","quant_sizes":{},"summary":"Antalia-2 Mini (version 1.0.1) is a small Turkish text-to-speech model with one voice. License: apache-2.0.","est":null,"license_class":"permissive"},{"type":"model","id":"mistralai/Voxtral-Mini-4B-Realtime-Arabic","name":"Voxtral-Mini-4B-Realtime-Arabic","org":"mistralai","date":"2026-10-08","license":"","params":4429679360,"params_src":"safetensors","active":null,"ctx":131072,"kv_bytes_tok":106496,"dtypes":["BF16"],"pipeline":"automatic-speech-recognition","modality":"audio","summary_what":"Voxtral Mini 4B Realtime Arabic is a streaming speech-to-text model for Arabic dialects and Modern Standard Arabic.","arch":{"class":"VoxtralRealtimeForConditionalGeneration","type":"dense"},"claims":["It outperforms several open-source and proprietary offline models and comes within 0.91 percentage points of Voxtral Transcribe Arabic\u2019s 7.91% CER."],"has_safetensors":true,"has_weights":true,"card_len":4994,"used_storage":17739214644,"created":"2026-10-08T11:36:06Z","first_seen":"2026-10-08T11:36:06Z","fresh":false,"url":"https://huggingface.co/mistralai/Voxtral-Mini-4B-Realtime-Arabic","likes":47,"downloads":1732,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":4.49,"lane":"main","quant_sizes":{"Q4":{"gb":2.2,"file":"Q4_K_M","repo":"mradermacher/Voxtral-Mini-4B-Realtime-Arabic-GGUF","kind":"GGUF","community":false},"Q8":{"gb":4.7,"file":"Q8_0","repo":"mradermacher/Voxtral-Mini-4B-Realtime-Arabic-GGUF","kind":"GGUF","community":false},"FP16":{"gb":8.9,"file":"F16","repo":"mradermacher/Voxtral-Mini-4B-Realtime-Arabic-GGUF","kind":"GGUF","community":false}},"summary":"Voxtral Mini 4B Realtime Arabic is a streaming speech-to-text model for Arabic dialects and Modern Standard Arabic. Dense, 4.4B params. Context: 128K tokens (config). Vendor-reported: \"It outperforms several open-source and proprietary offline models and comes within 0.91 percentage points of Voxtral Transcribe Arabic\u2019s 7.91% CER.\"","est":{"weights":{"Q4":2.2,"Q8":4.7,"FP16":8.9},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/Voxtral-Mini-4B-Realtime-Arabic-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/Voxtral-Mini-4B-Realtime-Arabic-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"mradermacher/Voxtral-Mini-4B-Realtime-Arabic-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.9,"kv_src":"config","need":{"Q4":4.1,"Q8":6.6,"FP16":10.8},"fit":{"nv8":"Q8","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"unknown"},{"type":"model","id":"mistralai/LIDstral-Arabic","name":"LIDstral-Arabic","org":"mistralai","date":"2026-10-08","license":"apache-2.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"text-classification","modality":"text","summary_what":"LIDstral Arabic is a fast language and dialect classifier for text written in Arabic script.","arch":{},"claims":[],"has_safetensors":false,"has_weights":true,"card_len":5695,"used_storage":1073874368,"created":"2026-10-08T11:37:05Z","first_seen":"2026-10-08T11:37:05Z","fresh":false,"url":"https://huggingface.co/mistralai/LIDstral-Arabic","likes":28,"downloads":173,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.51,"lane":"main","quant_sizes":{},"summary":"LIDstral Arabic is a fast language and dialect classifier for text written in Arabic script. License: apache-2.0.","est":null,"license_class":"permissive"},{"type":"model","id":"openbmb/MiniCPM-V-4.7-35B-A3B-DSpark","name":"MiniCPM-V-4.7-35B-A3B-DSpark","org":"openbmb","date":"2026-10-08","license":"apache-2.0","params":389223169,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":10240,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"Qwen3DSparkModel","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":4175,"used_storage":1556905716,"created":"2026-10-08T04:26:20Z","first_seen":"2026-10-08T04:26:20Z","fresh":false,"url":"https://huggingface.co/openbmb/MiniCPM-V-4.7-35B-A3B-DSpark","likes":26,"downloads":27563,"finetune":false,"gated":false,"quants":[{"id":"openbmb/MiniCPM-V-4.7-35B-A3B-DSpark-gguf","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":68.47,"lane":"main","quant_sizes":{"Q4":{"gb":0.2,"file":"Q4_K_M","repo":"openbmb/MiniCPM-V-4.7-35B-A3B-DSpark-gguf","kind":"GGUF","community":false},"Q8":{"gb":0.4,"file":"Q8_0","repo":"openbmb/MiniCPM-V-4.7-35B-A3B-DSpark-gguf","kind":"GGUF","community":false},"FP16":{"gb":0.8,"file":"F16","repo":"openbmb/MiniCPM-V-4.7-35B-A3B-DSpark-gguf","kind":"GGUF","community":false}},"summary":"Text generation model from openbmb (HF pipeline tag; the card has no one-line description). Dense, 389M params. Context: 256K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/minicpm-v","est":{"weights":{"Q4":0.2,"Q8":0.4,"FP16":0.8},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"openbmb/MiniCPM-V-4.7-35B-A3B-DSpark-gguf","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"openbmb/MiniCPM-V-4.7-35B-A3B-DSpark-gguf","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"openbmb/MiniCPM-V-4.7-35B-A3B-DSpark-gguf","fmt":"GGUF","community":false}},"kv_gb":0.1,"kv_src":"config","need":{"Q4":1.3,"Q8":1.5,"FP16":1.9},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"openbmb/MiniCPM-V-4.7-1B-DSpark","name":"MiniCPM-V-4.7-1B-DSpark","org":"openbmb","date":"2026-10-08","license":"apache-2.0","params":213547777,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":10240,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"Qwen3DSparkModel","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":3199,"used_storage":854204052,"created":"2026-10-08T04:26:19Z","first_seen":"2026-10-08T04:26:19Z","fresh":false,"url":"https://huggingface.co/openbmb/MiniCPM-V-4.7-1B-DSpark","likes":19,"downloads":15228,"finetune":false,"gated":false,"quants":[{"id":"openbmb/MiniCPM-V-4.7-1B-DSpark-gguf","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":37.84,"lane":"main","quant_sizes":{"Q4":{"gb":0.1,"file":"Q4_K_M","repo":"openbmb/MiniCPM-V-4.7-1B-DSpark-gguf","kind":"GGUF","community":false},"Q8":{"gb":0.2,"file":"Q8_0","repo":"openbmb/MiniCPM-V-4.7-1B-DSpark-gguf","kind":"GGUF","community":false},"FP16":{"gb":0.4,"file":"F16","repo":"openbmb/MiniCPM-V-4.7-1B-DSpark-gguf","kind":"GGUF","community":false}},"summary":"Text generation model from openbmb (HF pipeline tag; the card has no one-line description). Dense, 214M params. Context: 256K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/minicpm-v","est":{"weights":{"Q4":0.1,"Q8":0.2,"FP16":0.4},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"openbmb/MiniCPM-V-4.7-1B-DSpark-gguf","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"openbmb/MiniCPM-V-4.7-1B-DSpark-gguf","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"openbmb/MiniCPM-V-4.7-1B-DSpark-gguf","fmt":"GGUF","community":false}},"kv_gb":0.1,"kv_src":"config","need":{"Q4":1.2,"Q8":1.3,"FP16":1.5},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"nvidia/cosmos3-nano-libero10-sdg-export-S_best","name":"cosmos3-nano-libero10-sdg-export-S_best","org":"nvidia","date":"2026-10-08","license":"nvidia-open-model-license","params":15173136576,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"robotics","modality":"robotics","summary_what":"","arch":{"class":"Cosmos3OmniModel"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":1357,"used_storage":31499212848,"created":"2026-10-08T18:25:07Z","first_seen":"2026-10-18T23:26:24Z","fresh":false,"url":"https://huggingface.co/nvidia/cosmos3-nano-libero10-sdg-export-S_best","likes":0,"downloads":82,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.21,"lane":"main","quant_sizes":{"FP16":{"gb":31.5,"file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-S_best","kind":"safetensors","community":false}},"summary":"Robotics model from nvidia (HF pipeline tag; the card has no one-line description). 15B params. License: nvidia-open-model-license.","est":{"weights":{"Q4":9.1,"Q8":16.1,"FP16":31.5},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 15.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 15.2B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-S_best","fmt":"safetensors","community":false}},"kv_gb":1.3,"kv_src":"rough","need":{"Q4":11.4,"Q8":18.4,"FP16":33.8},"fit":{"nv8":null,"gpu12":null,"gpu16":"Q4","nv24":"Q8","nv32":"Q8","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q4","amd24":"Q8","strix":"FP16","gpu24":"Q8","gpu32":"Q8","mac16":"Q4","mac24":"Q4","mac36":"Q8","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q8","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"estimate","amd24":"estimate","strix":"real","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"nvidia/cosmos3-nano-libero10-sdg-export-S_worst","name":"cosmos3-nano-libero10-sdg-export-S_worst","org":"nvidia","date":"2026-10-08","license":"nvidia-open-model-license","params":15173136576,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"robotics","modality":"robotics","summary_what":"","arch":{"class":"Cosmos3OmniModel"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":1359,"used_storage":31499212848,"created":"2026-10-08T18:24:21Z","first_seen":"2026-10-18T23:26:24Z","fresh":false,"url":"https://huggingface.co/nvidia/cosmos3-nano-libero10-sdg-export-S_worst","likes":0,"downloads":78,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.2,"lane":"main","quant_sizes":{"FP16":{"gb":31.5,"file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-S_worst","kind":"safetensors","community":false}},"summary":"Robotics model from nvidia (HF pipeline tag; the card has no one-line description). 15B params. License: nvidia-open-model-license.","est":{"weights":{"Q4":9.1,"Q8":16.1,"FP16":31.5},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 15.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 15.2B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"nvidia/cosmos3-nano-libero10-sdg-export-S_worst","fmt":"safetensors","community":false}},"kv_gb":1.3,"kv_src":"rough","need":{"Q4":11.4,"Q8":18.4,"FP16":33.8},"fit":{"nv8":null,"gpu12":null,"gpu16":"Q4","nv24":"Q8","nv32":"Q8","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q4","amd24":"Q8","strix":"FP16","gpu24":"Q8","gpu32":"Q8","mac16":"Q4","mac24":"Q4","mac36":"Q8","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q8","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"estimate","amd24":"estimate","strix":"real","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"openbmb/MiniCPM-V-4.7-35B-A3B","name":"MiniCPM-V-4.7-35B-A3B","org":"openbmb","date":"2026-10-07","license":"apache-2.0","params":35212875824,"params_src":"safetensors","active":3000000000.0,"ctx":262144,"kv_bytes_tok":81920,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"MiniCPM-V 4.7 is the latest model in the MiniCPM-V series, available in a 35B-A3B variant, with 35B-A3B being the first MoE model in the MiniCPM-V series.","arch":{"class":"MiniCPMV4_7ForConditionalGeneration","type":"MoE","experts":256,"experts_active":8,"active":3000000000.0,"active_src":"name"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19842,"used_storage":70612796102,"created":"2026-10-07T07:07:30Z","first_seen":"2026-10-07T07:07:30Z","fresh":false,"url":"https://huggingface.co/openbmb/MiniCPM-V-4.7-35B-A3B","likes":1086,"downloads":187475,"finetune":false,"gated":false,"quants":[{"id":"bartowski/MiniCPM-V-4.7-35B-A3B-GGUF","kind":"GGUF"},{"id":"mlx-community/MiniCPM-V-4.7-35B-A3B-4bit","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-35B-A3B-5bit","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-35B-A3B-6bit","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-35B-A3B-8bit","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-35B-A3B-OptiQ-4bit","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-35B-A3B-mxfp4","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-35B-A3B-mxfp8","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-35B-A3B-nvfp4","kind":"MLX"},{"id":"openbmb/MiniCPM-V-4.7-35B-A3B-AWQ","kind":"quant"},{"id":"openbmb/MiniCPM-V-4.7-35B-A3B-FP8","kind":"quant"},{"id":"openbmb/MiniCPM-V-4.7-35B-A3B-GPTQ","kind":"quant"}],"x_mentions":[{"url":"https://x.com/grok/status/2111495391484977420","user":"grok","likes":0}],"x_likes":0,"x_posts":1,"signal":"trusted org","velocity":454.44,"lane":"main","quant_sizes":{"Q4":{"gb":21.2,"file":"Q4_K_M","repo":"openbmb/MiniCPM-V-4.7-35B-A3B-gguf","kind":"GGUF","community":false},"Q8":{"gb":36.9,"file":"Q8_0","repo":"openbmb/MiniCPM-V-4.7-35B-A3B-gguf","kind":"GGUF","community":false},"FP16":{"gb":70.2,"file":"F16","repo":"openbmb/MiniCPM-V-4.7-35B-A3B-gguf","kind":"GGUF","community":false}},"summary":"MiniCPM-V 4.7 is the latest model in the MiniCPM-V series, available in a 35B-A3B variant, with 35B-A3B being the first MoE model in the MiniCPM-V series. Mixture-of-experts, 35B total, 3B active (256 experts, 8 per token). Context: 256K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/minicpm-v","est":{"weights":{"Q4":21.2,"Q8":36.9,"FP16":70.2},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"openbmb/MiniCPM-V-4.7-35B-A3B-gguf","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"openbmb/MiniCPM-V-4.7-35B-A3B-gguf","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"openbmb/MiniCPM-V-4.7-35B-A3B-gguf","fmt":"GGUF","community":false}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":22.9,"Q8":38.6,"FP16":71.9},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q4","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"reflection/Beam-1.0-501B-A23B","name":"Beam-1.0-501B-A23B","org":"reflection","date":"2026-10-07","license":"apache-2.0","params":501267612352,"params_src":"safetensors","active":23000000000.0,"ctx":1048576,"kv_bytes_tok":212992,"dtypes":["BF16","F32"],"pipeline":"text-generation","modality":"text","summary_what":"Beam is an open model built by Reflection.","arch":{"class":"BeamForCausalLM","type":"MoE","experts":256,"experts_active":8,"active":23000000000.0,"active_src":"name"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19859,"used_storage":1002668503017,"created":"2026-10-07T19:52:01Z","first_seen":"2026-10-07T19:52:01Z","fresh":false,"url":"https://huggingface.co/reflection/Beam-1.0-501B-A23B","likes":254,"downloads":5209,"finetune":false,"gated":false,"quants":[{"id":"bartowski/Beam-1.0-501B-A23B-GGUF","kind":"GGUF"},{"id":"unsloth/Beam-1.0-501B-A23B-GGUF","kind":"GGUF"}],"x_mentions":[{"url":"https://x.com/astropol0/status/2111171994456875010","user":"astropol0","likes":0}],"x_likes":0,"x_posts":1,"signal":"trusted org","velocity":23.28,"lane":"main","quant_sizes":{"Q4":{"gb":306.5,"file":"UD-Q4_K_XL","repo":"unsloth/Beam-1.0-501B-A23B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":532.8,"file":"Q8_0","repo":"unsloth/Beam-1.0-501B-A23B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":1002.7,"file":"BF16","repo":"unsloth/Beam-1.0-501B-A23B-GGUF","kind":"GGUF","community":false}},"summary":"Beam is an open model built by Reflection. Mixture-of-experts, 501B total, 23B active (256 experts, 8 per token). Context: 1M tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/beam","est":{"weights":{"Q4":306.5,"Q8":532.8,"FP16":1002.7},"weights_src":{"Q4":{"kind":"real","file":"UD-Q4_K_XL","repo":"unsloth/Beam-1.0-501B-A23B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"unsloth/Beam-1.0-501B-A23B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"unsloth/Beam-1.0-501B-A23B-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.7,"kv_src":"config","need":{"Q4":309.2,"Q8":535.5,"FP16":1005.4},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":null,"amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":null,"mac512":"Q4","cpu32":null,"cpu64":null,"cpu128":null,"cpu256":null},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":null,"amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":null,"mac512":"real","cpu32":null,"cpu64":null,"cpu128":null,"cpu256":null},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"lightonai/LightOnOCR-3-0.8B","name":"LightOnOCR-3-0.8B","org":"lightonai","date":"2026-10-07","license":"apache-2.0","params":852985920,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":49152,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"LightOnOCR-3-0.8B is the lightest model of LightOn's LightOnOCR-3 family, with the same layout and visual understanding features.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":9022,"used_storage":1727668165,"created":"2026-10-07T10:23:52Z","first_seen":"2026-10-07T10:23:52Z","fresh":false,"url":"https://huggingface.co/lightonai/LightOnOCR-3-0.8B","likes":80,"downloads":7241,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"80 likes, 7241 downloads","velocity":17.39,"lane":"main","quant_sizes":{"Q4":{"gb":0.5,"file":"Q4_K_M","repo":"mradermacher/LightOnOCR-3-0.8B-i1-GGUF","kind":"GGUF","community":false},"Q8":{"gb":0.9,"file":"Q8_0","repo":"mradermacher/LightOnOCR-3-0.8B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":1.7,"file":"F16","repo":"mradermacher/LightOnOCR-3-0.8B-GGUF","kind":"GGUF","community":false}},"summary":"LightOnOCR-3-0.8B is the lightest model of LightOn's LightOnOCR-3 family, with the same layout and visual understanding features. Dense, 853M params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":0.5,"Q8":0.9,"FP16":1.7},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/LightOnOCR-3-0.8B-i1-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/LightOnOCR-3-0.8B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"mradermacher/LightOnOCR-3-0.8B-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.4,"kv_src":"config","need":{"Q4":1.9,"Q8":2.3,"FP16":3.1},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"ibm-granite/granite-speech-5.0-2b","name":"granite-speech-5.0-2b","org":"ibm-granite","date":"2026-10-07","license":"apache-2.0","params":2313207132,"params_src":"safetensors","active":null,"ctx":16384,"kv_bytes_tok":81920,"dtypes":["BF16"],"pipeline":"automatic-speech-recognition","modality":"audio","summary_what":"Granite Speech 5.0 2B is a compact and efficient speech-language model, specifically designed for multilingual automatic speech recognition (ASR) and bidirectional automatic speech translation (AST) for English, French, German, Spanish, Portuguese and Japanese.","arch":{"class":"GraniteSpeechForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19801,"used_storage":5040872787,"created":"2026-10-07T16:10:49Z","first_seen":"2026-10-21T23:23:07Z","fresh":false,"url":"https://huggingface.co/ibm-granite/granite-speech-5.0-2b","likes":49,"downloads":2377,"finetune":true,"gated":false,"quants":[{"id":"ibm-granite/granite-speech-5.0-2b-GGUF","kind":"GGUF"}],"x_mentions":[{"url":"https://x.com/bitsquin/status/2113227990951948552","user":"bitsquin","likes":0},{"url":"https://x.com/burnrate_ai/status/2113014511082844292","user":"burnrate_ai","likes":0}],"x_likes":0,"x_posts":2,"signal":"trusted org","velocity":25.84,"lane":"main","quant_sizes":{"Q4":{"gb":1.1,"file":"Q4_K_M","repo":"ibm-granite/granite-speech-5.0-2b-GGUF","kind":"GGUF","community":false},"Q8":{"gb":2.0,"file":"Q8_0","repo":"ibm-granite/granite-speech-5.0-2b-GGUF","kind":"GGUF","community":false},"FP16":{"gb":3.7,"file":"BF16","repo":"ibm-granite/granite-speech-5.0-2b-GGUF","kind":"GGUF","community":false}},"summary":"Granite Speech 5.0 2B is a compact and efficient speech-language model, specifically designed for multilingual automatic speech recognition (ASR) and bidirectional automatic speech translation (AST) for English, French, German, Spanish, Portuguese and Japanese. Dense, 2.3B params. Context: 16K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":1.1,"Q8":2.0,"FP16":3.7},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"ibm-granite/granite-speech-5.0-2b-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"ibm-granite/granite-speech-5.0-2b-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"ibm-granite/granite-speech-5.0-2b-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":2.8,"Q8":3.7,"FP16":5.4},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"nvidia/Nemotron-3.5-Embed-1B-BF16","name":"Nemotron-3.5-Embed-1B-BF16","org":"nvidia","date":"2026-10-07","license":"openmdw-1.1","params":1554710528,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":65536,"dtypes":["BF16"],"pipeline":"visual-document-retrieval","modality":"embedding","summary_what":"The Nemotron-3.5-Embed-1B-BF16 model was distilled from the Nemotron-3.5-Embed-8B-BF16 embedding teacher using combined cosine distance (COS) and mean squared error (MSE) losses on multilingual text and image retrieval data.","arch":{"class":"Mistral3Model","type":"dense"},"claims":["Among models of comparable size, Nemotron-3.5-Embed-1B-BF16 achieves state-of-the-art results across multiple multilingual and multimodal retrieval benchmarks."],"has_safetensors":true,"has_weights":true,"card_len":19367,"used_storage":3130994338,"created":"2026-10-07T10:46:18Z","first_seen":"2026-10-07T10:46:18Z","fresh":false,"url":"https://huggingface.co/nvidia/Nemotron-3.5-Embed-1B-BF16","likes":44,"downloads":4669,"finetune":true,"gated":false,"quants":[{"id":"nvidia/Nemotron-3.5-Embed-1B-NVFP4","kind":"quant"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":11.2,"lane":"main","quant_sizes":{"Q4":{"gb":0.7,"file":"Q4_K_M","repo":"W4D/Nemotron-3.5-Embed-1B-GGUF","kind":"GGUF","community":true},"Q8":{"gb":1.2,"file":"Q8_0","repo":"W4D/Nemotron-3.5-Embed-1B-GGUF","kind":"GGUF","community":true},"FP16":{"gb":3.1,"file":"BF16","repo":"W4D/Nemotron-3.5-Embed-1B-GGUF","kind":"GGUF","community":true}},"summary":"The Nemotron-3.5-Embed-1B-BF16 model was distilled from the Nemotron-3.5-Embed-8B-BF16 embedding teacher using combined cosine distance (COS) and mean squared error (MSE) losses on multilingual text and image retrieval data. Dense, 1.6B params. Context: 256K tokens (config). License: openmdw-1.1. Vendor-reported: \"Among models of comparable size, Nemotron-3.5-Embed-1B-BF16 achieves state-of-the-art results across multiple multilingual and multimodal retrieval benchmarks.\"","ollama":"https://ollama.com/library/d1","est":{"weights":{"Q4":0.7,"Q8":1.2,"FP16":3.1},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"W4D/Nemotron-3.5-Embed-1B-GGUF","fmt":"GGUF","community":true},"Q8":{"kind":"real","file":"Q8_0","repo":"W4D/Nemotron-3.5-Embed-1B-GGUF","fmt":"GGUF","community":true},"FP16":{"kind":"real","file":"BF16","repo":"W4D/Nemotron-3.5-Embed-1B-GGUF","fmt":"GGUF","community":true}},"kv_gb":0.5,"kv_src":"config","need":{"Q4":2.2,"Q8":2.7,"FP16":4.6},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"nvidia/Nemotron-3.5-Embed-8B-BF16","name":"Nemotron-3.5-Embed-8B-BF16","org":"nvidia","date":"2026-10-07","license":"openmdw-1.1","params":8381155328,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":139264,"dtypes":["BF16"],"pipeline":"visual-document-retrieval","modality":"embedding","summary_what":"","arch":{"class":"Mistral3Model","type":"dense"},"claims":["Nemotron-3.5-Embed-8B-BF16 achieves state-of-the-art results across multiple multilingual and multimodal retrieval benchmarks."],"has_safetensors":true,"has_weights":true,"card_len":19367,"used_storage":16783904442,"created":"2026-10-07T13:12:32Z","first_seen":"2026-10-07T13:12:32Z","fresh":false,"url":"https://huggingface.co/nvidia/Nemotron-3.5-Embed-8B-BF16","likes":29,"downloads":2273,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":5.5,"lane":"main","quant_sizes":{"Q4":{"gb":4.9,"file":"Q4_K_M","repo":"W4D/Nemotron-3.5-Embed-8B-GGUF","kind":"GGUF","community":true},"Q8":{"gb":8.5,"file":"Q8_0","repo":"W4D/Nemotron-3.5-Embed-8B-GGUF","kind":"GGUF","community":true},"FP16":{"gb":16.8,"file":"BF16","repo":"W4D/Nemotron-3.5-Embed-8B-GGUF","kind":"GGUF","community":true}},"summary":"Visual document retrieval model from nvidia (HF pipeline tag; the card has no one-line description). Dense, 8.4B params. Context: 256K tokens (config). License: openmdw-1.1. Vendor-reported: \"Nemotron-3.5-Embed-8B-BF16 achieves state-of-the-art results across multiple multilingual and multimodal retrieval benchmarks.\"","ollama":"https://ollama.com/library/nemotron3","est":{"weights":{"Q4":4.9,"Q8":8.5,"FP16":16.8},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"W4D/Nemotron-3.5-Embed-8B-GGUF","fmt":"GGUF","community":true},"Q8":{"kind":"real","file":"Q8_0","repo":"W4D/Nemotron-3.5-Embed-8B-GGUF","fmt":"GGUF","community":true},"FP16":{"kind":"real","file":"BF16","repo":"W4D/Nemotron-3.5-Embed-8B-GGUF","fmt":"GGUF","community":true}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":7.0,"Q8":10.6,"FP16":18.9},"fit":{"nv8":"Q4","gpu12":"Q8","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"Q8","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"openbmb/MiniCPM-V-4.7-1B","name":"MiniCPM-V-4.7-1B","org":"openbmb","date":"2026-10-06","license":"apache-2.0","params":1554476272,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":49152,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"MiniCPM-V 4.7 is the latest model in the MiniCPM-V series, available in a 1B variant.","arch":{"class":"MiniCPMV4_7ForConditionalGeneration","type":"dense"},"claims":["In Instruct mode, the 1B variant continues to improve foundation capabilities over MiniCPM-V 4.6 and substantially outperforms Qwen3.5 0.8B in both overall score and inference efficiency.","In Instruct mode, 1B substantially outperforms Qwen3.5 0.8B in both overall score and inference efficiency."],"has_safetensors":true,"has_weights":true,"card_len":19848,"used_storage":3295937680,"created":"2026-10-06T17:40:13Z","first_seen":"2026-10-06T17:40:13Z","fresh":false,"url":"https://huggingface.co/openbmb/MiniCPM-V-4.7-1B","likes":405,"downloads":104433,"finetune":false,"gated":false,"quants":[{"id":"mlx-community/MiniCPM-V-4.7-1B-4bit","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-1B-5bit","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-1B-6bit","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-1B-8bit","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-1B-OptiQ-4bit","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-1B-mxfp4","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-1B-mxfp8","kind":"MLX"},{"id":"mlx-community/MiniCPM-V-4.7-1B-nvfp4","kind":"MLX"},{"id":"openbmb/MiniCPM-V-4.7-1B-AWQ","kind":"quant"},{"id":"openbmb/MiniCPM-V-4.7-1B-GPTQ","kind":"quant"},{"id":"openbmb/MiniCPM-V-4.7-1B-gguf","kind":"GGUF"},{"id":"openbmb/MiniCPM-V-4.7-1B-int4","kind":"quant"}],"x_mentions":[{"url":"https://x.com/CleoWal84405014/status/2110422817628172339","user":"CleoWal84405014","likes":0}],"x_likes":0,"x_posts":1,"signal":"trusted org","velocity":249.51,"lane":"main","quant_sizes":{"Q4":{"gb":0.7,"file":"Q4_K_M","repo":"openbmb/MiniCPM-V-4.7-1B-gguf","kind":"GGUF","community":false},"Q8":{"gb":1.1,"file":"Q8_0","repo":"openbmb/MiniCPM-V-4.7-1B-gguf","kind":"GGUF","community":false},"FP16":{"gb":2.5,"file":"F16","repo":"openbmb/MiniCPM-V-4.7-1B-gguf","kind":"GGUF","community":false}},"summary":"MiniCPM-V 4.7 is the latest model in the MiniCPM-V series, available in a 1B variant. Dense, 1.6B params. Context: 256K tokens (config). License: apache-2.0. Vendor-reported: \"In Instruct mode, the 1B variant continues to improve foundation capabilities over MiniCPM-V 4.6 and substantially outperforms Qwen3.5 0.8B in both overall score and inference efficiency.\" Vendor-reported: \"In Instruct mode, 1B substantially outperforms Qwen3.5 0.8B in both overall score and inference efficiency.\"","ollama":"https://ollama.com/library/minicpm-v","est":{"weights":{"Q4":0.7,"Q8":1.1,"FP16":2.5},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"openbmb/MiniCPM-V-4.7-1B-gguf","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"openbmb/MiniCPM-V-4.7-1B-gguf","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"openbmb/MiniCPM-V-4.7-1B-gguf","fmt":"GGUF","community":false}},"kv_gb":0.4,"kv_src":"config","need":{"Q4":2.1,"Q8":2.5,"FP16":3.9},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"lightonai/LightOnOCR-3-4B","name":"LightOnOCR-3-4B","org":"lightonai","date":"2026-10-06","license":"apache-2.0","params":4539265536,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":131072,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"LightOnOCR-3-4B is the largest and most accurate model of the LightOnOCR-3 family, LightOn's new generation of lightweight end-to-end OCR models with layout and visual understanding.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":9095,"used_storage":9100258191,"created":"2026-10-06T16:12:40Z","first_seen":"2026-10-06T16:12:40Z","fresh":false,"url":"https://huggingface.co/lightonai/LightOnOCR-3-4B","likes":113,"downloads":13866,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"113 likes, 13866 downloads","velocity":31.83,"lane":"main","quant_sizes":{"Q4":{"gb":2.7,"file":"Q4_K_M","repo":"mradermacher/LightOnOCR-3-4B-i1-GGUF","kind":"GGUF","community":false},"Q8":{"gb":4.8,"file":"Q8_0","repo":"mradermacher/LightOnOCR-3-4B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":9.1,"file":"F16","repo":"mradermacher/LightOnOCR-3-4B-GGUF","kind":"GGUF","community":false}},"summary":"LightOnOCR-3-4B is the largest and most accurate model of the LightOnOCR-3 family, LightOn's new generation of lightweight end-to-end OCR models with layout and visual understanding. Dense, 4.5B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":2.7,"Q8":4.8,"FP16":9.1},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/LightOnOCR-3-4B-i1-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/LightOnOCR-3-4B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"mradermacher/LightOnOCR-3-4B-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":4.8,"Q8":6.9,"FP16":11.2},"fit":{"nv8":"Q8","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"nvidia/RFD4-Proteina","name":"RFD4-Proteina","org":"nvidia","date":"2026-10-06","license":"bsd-3-clause","params":435622465,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"","modality":"","summary_what":"Use the RFD4 loader below; this is a native RFD4 bundle.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19784,"used_storage":178077620620,"created":"2026-10-06T05:00:39Z","first_seen":"2026-10-21T23:23:07Z","fresh":false,"url":"https://huggingface.co/nvidia/RFD4-Proteina","likes":26,"downloads":3149,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":7.05,"lane":"main","quant_sizes":{},"summary":"Use the RFD4 loader below; this is a native RFD4 bundle. 436M params. License: bsd-3-clause.","est":{"weights":{"Q4":0.3,"Q8":0.5,"FP16":0.9},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.4B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.4B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.4B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.8,"Q8":2.0,"FP16":2.4},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"LiquidAI/d1-3B","name":"d1-3B","org":"LiquidAI","date":"2026-10-05","license":"lfm1.0","params":3123483888,"params_src":"safetensors","active":null,"ctx":32768,"kv_bytes_tok":61440,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"d1-3B is a 3B parameter decision model built on LFM2.5-VL-3B.","arch":{"class":"Lfm2VlForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":10057,"used_storage":12574176108,"created":"2026-10-05T09:55:28Z","first_seen":"2026-10-05T09:55:28Z","fresh":false,"url":"https://huggingface.co/LiquidAI/d1-3B","likes":477,"downloads":26547,"finetune":true,"gated":false,"quants":[{"id":"LiquidAI/d1-3B-GGUF","kind":"GGUF"},{"id":"LiquidAI/d1-3B-MLX-4bit","kind":"MLX"},{"id":"LiquidAI/d1-3B-MLX-8bit","kind":"MLX"},{"id":"LiquidAI/d1-3B-MLX-fp16","kind":"MLX"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":57.56,"lane":"main","quant_sizes":{"Q4":{"gb":1.7,"file":"Q4_K_M","repo":"LiquidAI/d1-3B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":2.9,"file":"Q8_0","repo":"LiquidAI/d1-3B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":5.4,"file":"BF16","repo":"LiquidAI/d1-3B-GGUF","kind":"GGUF","community":false}},"summary":"d1-3B is a 3B parameter decision model built on LFM2.5-VL-3B. Dense, 3.1B params. Context: 32K tokens (config). License: lfm1.0.","ollama":"https://ollama.com/library/d1","est":{"weights":{"Q4":1.7,"Q8":2.9,"FP16":5.4},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"LiquidAI/d1-3B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"LiquidAI/d1-3B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"LiquidAI/d1-3B-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.5,"kv_src":"config","need":{"Q4":3.2,"Q8":4.4,"FP16":6.9},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"restricted"},{"type":"model","id":"speridlabs/iris-3b","name":"iris-3b","org":"speridlabs","date":"2026-10-05","license":"apache-2.0","params":2987511888,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"text-to-image","modality":"image","summary_what":"Iris-3B Turbo, in the turbo/ folder of this repo, is a 4-step distilled version of Iris-3B.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":15400,"used_storage":47887825955,"created":"2026-10-05T14:51:15Z","first_seen":"2026-10-05T14:51:15Z","fresh":false,"url":"https://huggingface.co/speridlabs/iris-3b","likes":328,"downloads":1612,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"328 likes","velocity":4.18,"lane":"main","quant_sizes":{"Q4":{"gb":1.7,"file":"Q4_K_S","repo":"realrebelai/iris-3b_GGUFs","kind":"GGUF","community":true}},"summary":"Iris-3B Turbo, in the turbo/ folder of this repo, is a 4-step distilled version of Iris-3B. 3B params. License: apache-2.0.","est":{"weights":{"Q4":1.7,"Q8":3.2,"FP16":6.0},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_S","repo":"realrebelai/iris-3b_GGUFs","fmt":"GGUF","community":true},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 3.0B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 3.0B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":3.2,"Q8":4.7,"FP16":7.5},"fit":{"nv8":"Q8","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"perplexity-ai/pplx-decider-v1.1-27b","name":"pplx-decider-v1.1-27b","org":"perplexity-ai","date":"2026-10-05","license":"apache-2.0","params":26085330160,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16"],"pipeline":"text-classification","modality":"text","summary_what":"pplx-decider-v1.1-27b is an update of our pplx-decider-v1-27b model, that boost the Decision Index score from 56.4 to 61.56, now outperforming Jev by more than 3.5 points, while using the same backbone.","arch":{"class":"Qwen3_5Model","type":"dense"},"claims":["pplx-decider-v1.1-27b is an update of our pplx-decider-v1-27b model, that boost the Decision Index score from 56.4 to 61.56, now outperforming Jev by more than 3.5 points, while using the same backbone."],"has_safetensors":true,"has_weights":true,"card_len":6471,"used_storage":52193403420,"created":"2026-10-05T12:32:27Z","first_seen":"2026-10-05T12:32:27Z","fresh":false,"url":"https://huggingface.co/perplexity-ai/pplx-decider-v1.1-27b","likes":167,"downloads":5106,"finetune":true,"gated":false,"quants":[{"id":"mlx-community/pplx-decider-v1.1-27b-4bit","kind":"MLX"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":11.29,"lane":"main","quant_sizes":{"Q8":{"gb":27.2,"file":"Q8_0","repo":"Livesport/pplx-decider-v1.1-27b-GGUF","kind":"GGUF","community":true},"Q4":{"gb":15.3,"file":"MLX 4-bit","repo":"mlx-community/pplx-decider-v1.1-27b-4bit","kind":"MLX","community":false},"FP16":{"gb":52.2,"file":"official BF16 safetensors","repo":"perplexity-ai/pplx-decider-v1.1-27b","kind":"safetensors","community":false}},"summary":"pplx-decider-v1.1-27b is an update of our pplx-decider-v1-27b model, that boost the Decision Index score from 56.4 to 61.56, now outperforming Jev by more than 3.5 points, while using the same backbone. Dense, 26B params. Context: 256K tokens (config). License: apache-2.0. Vendor-reported: \"pplx-decider-v1.1-27b is an update of our pplx-decider-v1-27b model, that boost the Decision Index score from 56.4 to 61.56, now outperforming Jev by more than 3.5 points, while using the same backbone.\"","ollama":"https://ollama.com/library/pplx-decider","est":{"weights":{"Q4":15.3,"Q8":27.2,"FP16":52.2},"weights_src":{"Q4":{"kind":"real","file":"MLX 4-bit","repo":"mlx-community/pplx-decider-v1.1-27b-4bit","fmt":"MLX","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"Livesport/pplx-decider-v1.1-27b-GGUF","fmt":"GGUF","community":true},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"perplexity-ai/pplx-decider-v1.1-27b","fmt":"safetensors","community":false}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":18.4,"Q8":30.3,"FP16":55.3},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q8","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":"Q8","mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":"real","mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"LiquidAI/d1-omni-600M","name":"d1-omni-600M","org":"LiquidAI","date":"2026-10-05","license":"lfm1.0","params":587161089,"params_src":"safetensors","active":null,"ctx":128000,"kv_bytes_tok":32768,"dtypes":["F32"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"d1-omni-600M is a 600M parameter decision model built on LFM2.5-Encoder-350M.","arch":{"class":"D1OmniModel","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":9974,"used_storage":4718932608,"created":"2026-10-05T17:18:48Z","first_seen":"2026-10-05T17:18:48Z","fresh":false,"url":"https://huggingface.co/LiquidAI/d1-omni-600M","likes":141,"downloads":17430,"finetune":true,"gated":false,"quants":[{"id":"LiquidAI/d1-omni-600M-GGUF","kind":"GGUF"},{"id":"LiquidAI/d1-omni-600M-MLX-8bit","kind":"MLX"},{"id":"LiquidAI/d1-omni-600M-MLX-fp16","kind":"MLX"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":38.03,"lane":"main","quant_sizes":{"Q8":{"gb":0.4,"file":"Q8_0","repo":"LiquidAI/d1-omni-600M-GGUF","kind":"GGUF","community":false},"FP16":{"gb":0.8,"file":"BF16","repo":"LiquidAI/d1-omni-600M-GGUF","kind":"GGUF","community":false},"Q4":{"gb":0.2,"file":"Q4_K_M","repo":"mradermacher/d1-omni-600M-GGUF","kind":"GGUF","community":false}},"summary":"d1-omni-600M is a 600M parameter decision model built on LFM2.5-Encoder-350M. Dense, 587M params. Context: 125K tokens (config). License: lfm1.0.","ollama":"https://ollama.com/library/d1-omni","est":{"weights":{"Q4":0.2,"Q8":0.4,"FP16":0.8},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/d1-omni-600M-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"LiquidAI/d1-omni-600M-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"LiquidAI/d1-omni-600M-GGUF","fmt":"GGUF","community":false}},"kv_gb":0.3,"kv_src":"config","need":{"Q4":1.5,"Q8":1.7,"FP16":2.1},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"restricted"},{"type":"model","id":"inclusionAI/LemonDecider","name":"LemonDecider","org":"inclusionAI","date":"2026-10-05","license":"apache-2.0","params":4539918336,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":131072,"dtypes":["BF16","F32"],"pipeline":"feature-extraction","modality":"embedding","summary_what":"LemonDecider 4B is a typed decision engine: it reads a state and a question with a list of options, runs one forward pass, and returns a probability distribution over the options (choice / noul questions) or an ordered rating (score questions).","arch":{"class":"LemonDeciderForDecisionClassification","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":13906,"used_storage":18181151636,"created":"2026-10-05T04:27:12Z","first_seen":"2026-10-05T04:27:12Z","fresh":false,"url":"https://huggingface.co/inclusionAI/LemonDecider","likes":33,"downloads":504,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":1.13,"lane":"main","quant_sizes":{"FP16":{"gb":9.1,"file":"official BF16 safetensors","repo":"inclusionAI/LemonDecider","kind":"safetensors","community":false}},"summary":"LemonDecider 4B is a typed decision engine: it reads a state and a question with a list of options, runs one forward pass, and returns a probability distribution over the options (choice / noul questions) or an ordered rating (score questions). Dense, 4.5B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":2.7,"Q8":4.8,"FP16":9.1},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 4.5B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 4.5B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"inclusionAI/LemonDecider","fmt":"safetensors","community":false}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":4.8,"Q8":6.9,"FP16":11.2},"fit":{"nv8":"Q8","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"amazon/ALoDLM-8B","name":"ALoDLM-8B","org":"amazon","date":"2026-10-02","license":"cc-by-nc-4.0","params":8190735360,"params_src":"safetensors","active":null,"ctx":40960,"kv_bytes_tok":147456,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"ALoDLM is a family of diffusion language models developed at Amazon, available as ALoDLM-1.7B and ALoDLM-8B and initialized from the corresponding Qwen3 backbones.","arch":{"class":"Qwen3ForCausalLM","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":13448,"used_storage":16394454899,"created":"2026-10-02T18:46:26Z","first_seen":"2026-10-02T18:46:26Z","fresh":false,"url":"https://huggingface.co/amazon/ALoDLM-8B","likes":62,"downloads":2016,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":3.9,"lane":"main","quant_sizes":{"Q4":{"gb":5.0,"file":"Q4_K_M","repo":"mradermacher/ALoDLM-8B-i1-GGUF","kind":"GGUF","community":false},"Q8":{"gb":8.7,"file":"Q8_0","repo":"mradermacher/ALoDLM-8B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":16.4,"file":"F16","repo":"mradermacher/ALoDLM-8B-GGUF","kind":"GGUF","community":false}},"summary":"ALoDLM is a family of diffusion language models developed at Amazon, available as ALoDLM-1.7B and ALoDLM-8B and initialized from the corresponding Qwen3 backbones. Dense, 8.2B params. Context: 40K tokens (config). License: cc-by-nc-4.0.","est":{"weights":{"Q4":5.0,"Q8":8.7,"FP16":16.4},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"mradermacher/ALoDLM-8B-i1-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"mradermacher/ALoDLM-8B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"F16","repo":"mradermacher/ALoDLM-8B-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.2,"kv_src":"config","need":{"Q4":7.2,"Q8":10.9,"FP16":18.6},"fit":{"nv8":null,"gpu12":"Q8","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"Q8","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"restricted"},{"type":"model","id":"SoMarkAI/OXR-1.0","name":"OXR-1.0","org":"SoMarkAI","date":"2026-10-02","license":"","params":1534378216,"params_src":"safetensors","active":null,"ctx":40960,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"","modality":"","summary_what":"OXR (Optical Everything Recognition) is an open-source model from the SoMark team.","arch":{"class":"OXR","type":"dense"},"claims":["An open-source document parsing model with SOTA accuracy and speed, using an RSI loop during post-training.","With just 1.3B parameters, OXR ranks #1 on olmOCR-Bench, OmniDocBench, and DocParsingBench while delivering industry-leading parsing speed."],"has_safetensors":true,"has_weights":true,"card_len":19558,"used_storage":3159071481,"created":"2026-10-02T13:30:18Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/SoMarkAI/OXR-1.0","likes":53,"downloads":528,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"53 likes","velocity":1.08,"lane":"main","quant_sizes":{"FP16":{"gb":3.1,"file":"official BF16 safetensors","repo":"SoMarkAI/OXR-1.0","kind":"safetensors","community":false}},"summary":"OXR (Optical Everything Recognition) is an open-source model from the SoMark team. Dense, 1.5B params. Context: 40K tokens (config). Vendor-reported: \"An open-source document parsing model with SOTA accuracy and speed, using an RSI loop during post-training.\" Vendor-reported: \"With just 1.3B parameters, OXR ranks #1 on olmOCR-Bench, OmniDocBench, and DocParsingBench while delivering industry-leading parsing speed.\"","est":{"weights":{"Q4":0.9,"Q8":1.6,"FP16":3.1},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 1.5B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 1.5B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"SoMarkAI/OXR-1.0","fmt":"safetensors","community":false}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":2.4,"Q8":3.1,"FP16":4.6},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"unknown"},{"type":"model","id":"nvidia/Flora-Rewarder","name":"Flora-Rewarder","org":"nvidia","date":"2026-10-02","license":"nvidia-license","params":27781428976,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16","F32","U8"],"pipeline":"video-text-to-text","modality":"vision","summary_what":"Training labels are a combination of human-written annotations, automatic success signals recorded with the data, and VLM-generated annotations.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":9940,"used_storage":111146137305,"created":"2026-10-02T18:01:10Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/nvidia/Flora-Rewarder","likes":17,"downloads":979,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":1.87,"lane":"main","quant_sizes":{},"summary":"Training labels are a combination of human-written annotations, automatic success signals recorded with the data, and VLM-generated annotations. Dense, 28B params. Context: 256K tokens (config). License: nvidia-license.","est":{"weights":{"Q4":16.7,"Q8":29.5,"FP16":55.6},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 27.8B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 27.8B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 27.8B params"}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":19.8,"Q8":32.6,"FP16":58.7},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":null,"amd24":"estimate","strix":"estimate","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"amazon/ALoDLM-1.7B","name":"ALoDLM-1.7B","org":"amazon","date":"2026-10-02","license":"cc-by-nc-4.0","params":2031739904,"params_src":"safetensors","active":null,"ctx":40960,"kv_bytes_tok":114688,"dtypes":["BF16"],"pipeline":"text-generation","modality":"text","summary_what":"ALoDLM is a family of diffusion language models developed at Amazon, available as ALoDLM-1.7B and ALoDLM-8B and initialized from the corresponding Qwen3 backbones.","arch":{"class":"Qwen3ForCausalLM","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":13263,"used_storage":4076449619,"created":"2026-10-02T18:46:55Z","first_seen":"2026-10-02T18:46:55Z","fresh":false,"url":"https://huggingface.co/amazon/ALoDLM-1.7B","likes":12,"downloads":517,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.99,"lane":"main","quant_sizes":{"Q8":{"gb":2.2,"file":"Q8_0","repo":"wahidmounir/ALoDLM-1.7B-Q8_0-GGUF","kind":"GGUF","community":true},"FP16":{"gb":4.1,"file":"official BF16 safetensors","repo":"amazon/ALoDLM-1.7B","kind":"safetensors","community":false}},"summary":"ALoDLM is a family of diffusion language models developed at Amazon, available as ALoDLM-1.7B and ALoDLM-8B and initialized from the corresponding Qwen3 backbones. Dense, 2B params. Context: 40K tokens (config). License: cc-by-nc-4.0.","est":{"weights":{"Q4":1.2,"Q8":2.2,"FP16":4.1},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 2.0B params"},"Q8":{"kind":"real","file":"Q8_0","repo":"wahidmounir/ALoDLM-1.7B-Q8_0-GGUF","fmt":"GGUF","community":true},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"amazon/ALoDLM-1.7B","fmt":"safetensors","community":false}},"kv_gb":0.9,"kv_src":"config","need":{"Q4":3.1,"Q8":4.1,"FP16":6.0},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"fishaudio/drama-3","name":"drama-3","org":"fishaudio","date":"2026-10-01","license":"fish-audio-research-license","params":35899643008,"params_src":"safetensors","active":3000000000.0,"ctx":262144,"kv_bytes_tok":81920,"dtypes":["BF16","F8_E4M3"],"pipeline":"text-to-speech","modality":"audio","summary_what":"Fish Audio Drama 3 is Fish Audio's new-generation text-to-speech model, built for multi-character scenes and long-form content.","arch":{"class":"FishOmniModel","type":"MoE","experts":256,"experts_active":8,"active":3000000000.0,"active_src":"model card"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":7164,"used_storage":41495465142,"created":"2026-10-01T04:34:57Z","first_seen":"2026-10-01T04:34:57Z","fresh":false,"url":"https://huggingface.co/fishaudio/drama-3","likes":979,"downloads":19831,"finetune":false,"gated":false,"quants":[{"id":"fishaudio/drama-3-int4","kind":"quant"},{"id":"mlx-community/fish-audio-drama-3-3bit","kind":"MLX"},{"id":"mlx-community/fish-audio-drama-3-4bit","kind":"MLX"},{"id":"mlx-community/fish-audio-drama-3-8bit","kind":"MLX"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"979 likes, 19831 downloads","velocity":36.46,"lane":"main","quant_sizes":{"Q8":{"gb":38.5,"file":"Q8_0","repo":"sakasegawa/drama-3-GGUF","kind":"GGUF","community":true},"FP16":{"gb":71.6,"file":"F16","repo":"sakasegawa/drama-3-GGUF","kind":"GGUF","community":true},"Q4":{"gb":23.8,"file":"MLX 4-bit","repo":"mlx-community/fish-audio-drama-3-4bit","kind":"MLX","community":false}},"summary":"Fish Audio Drama 3 is Fish Audio's new-generation text-to-speech model, built for multi-character scenes and long-form content. Mixture-of-experts, 36B total, 3B active (256 experts, 8 per token). Context: 256K tokens (config). License: fish-audio-research-license.","est":{"weights":{"Q4":23.8,"Q8":38.5,"FP16":71.6},"weights_src":{"Q4":{"kind":"real","file":"MLX 4-bit","repo":"mlx-community/fish-audio-drama-3-4bit","fmt":"MLX","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"sakasegawa/drama-3-GGUF","fmt":"GGUF","community":true},"FP16":{"kind":"real","file":"F16","repo":"sakasegawa/drama-3-GGUF","fmt":"GGUF","community":true}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":25.5,"Q8":40.2,"FP16":73.3},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":null,"strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q4","mac64":"Q8","mac96":"Q8","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"restricted"},{"type":"model","id":"canberkkkkkk/ema-lightning","name":"ema-lightning","org":"canberkkkkkk","date":"2026-10-01","license":"apache-2.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"text-to-speech","modality":"audio","summary_what":"","arch":{},"claims":[],"has_safetensors":false,"has_weights":true,"card_len":17575,"used_storage":36738216,"created":"2026-10-01T14:18:49Z","first_seen":"2026-10-01T14:18:49Z","fresh":false,"url":"https://huggingface.co/canberkkkkkk/ema-lightning","likes":444,"downloads":59700,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"444 likes, 59700 downloads","velocity":107.19,"lane":"main","quant_sizes":{},"summary":"Text to speech model from canberkkkkkk (HF pipeline tag; the card has no one-line description). License: apache-2.0.","est":null,"license_class":"permissive"},{"type":"model","id":"perplexity-ai/pplx-decider-v1-27b","name":"pplx-decider-v1-27b","org":"perplexity-ai","date":"2026-10-01","license":"apache-2.0","params":26085330160,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16"],"pipeline":"text-classification","modality":"text","summary_what":"pplx-decider-v1-27b is a decision model fine-tuned from Qwen3.8-27B.","arch":{"class":"Qwen3_5Model","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":2966,"used_storage":52193403420,"created":"2026-10-01T21:24:11Z","first_seen":"2026-10-01T21:24:11Z","fresh":false,"url":"https://huggingface.co/perplexity-ai/pplx-decider-v1-27b","likes":132,"downloads":3508,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":6.57,"lane":"main","quant_sizes":{"FP16":{"gb":52.2,"file":"official BF16 safetensors","repo":"perplexity-ai/pplx-decider-v1-27b","kind":"safetensors","community":false}},"summary":"pplx-decider-v1-27b is a decision model fine-tuned from Qwen3.8-27B. Dense, 26B params. Context: 256K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/pplx-decider","est":{"weights":{"Q4":15.7,"Q8":27.7,"FP16":52.2},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 26.1B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 26.1B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"perplexity-ai/pplx-decider-v1-27b","fmt":"safetensors","community":false}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":18.8,"Q8":30.8,"FP16":55.3},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q8","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":"Q8","mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"estimate","strix":"real","gpu24":null,"gpu32":"estimate","mac16":null,"mac24":null,"mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"nvidia/PixelUMM","name":"PixelUMM","org":"nvidia","date":"2026-10-01","license":"nvidia-one-way-noncommercial-license","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"any-to-any","modality":"omni","summary_what":"PixelUMM is an NVIDIA-developed, encoder-free unified multimodal model for joint understanding and generation across text, images, and video directly in pixel space.","arch":{},"claims":[],"has_safetensors":false,"has_weights":false,"card_len":3832,"used_storage":121628326163,"created":"2026-10-01T21:40:34Z","first_seen":"2026-10-01T21:40:34Z","fresh":false,"url":"https://huggingface.co/nvidia/PixelUMM","likes":89,"downloads":0,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.16,"lane":"main","quant_sizes":{},"summary":"PixelUMM is an NVIDIA-developed, encoder-free unified multimodal model for joint understanding and generation across text, images, and video directly in pixel space. License: nvidia-one-way-noncommercial-license.","est":null,"license_class":"restricted"},{"type":"model","id":"nvidia/Flora-Captioner","name":"Flora-Captioner","org":"nvidia","date":"2026-10-01","license":"nvidia-license","params":27356728560,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["F32","U8"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":13859,"used_storage":164173512290,"created":"2026-10-01T22:22:57Z","first_seen":"2026-10-20T23:23:28Z","fresh":false,"url":"https://huggingface.co/nvidia/Flora-Captioner","likes":34,"downloads":1134,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":2.11,"lane":"main","quant_sizes":{},"summary":"Image text to text model from nvidia (HF pipeline tag; the card has no one-line description). Dense, 27B params. Context: 256K tokens (config). License: nvidia-license.","est":{"weights":{"Q4":16.4,"Q8":29.1,"FP16":54.7},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 27.4B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 27.4B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 27.4B params"}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":19.5,"Q8":32.2,"FP16":57.8},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":null,"amd24":"estimate","strix":"estimate","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"restricted"},{"type":"model","id":"ibm-granite/ensemble-forecasting-with-ibm-granite-time-series","name":"ensemble-forecasting-with-ibm-granite-time-series","org":"ibm-granite","date":"2026-10-01","license":"","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"","modality":"","summary_what":"","arch":{},"claims":[],"has_safetensors":false,"has_weights":false,"card_len":0,"used_storage":0,"created":"2026-10-01T16:28:23Z","first_seen":"2026-10-01T16:28:23Z","fresh":false,"url":"https://huggingface.co/ibm-granite/ensemble-forecasting-with-ibm-granite-time-series","likes":2,"downloads":0,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.0,"lane":"main","quant_sizes":{},"summary":"","est":null,"license_class":"unknown"},{"type":"model","id":"Cloudflare/clef","name":"clef","org":"Cloudflare","date":"2026-09-30","license":"apache-2.0","params":27356728560,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"Clef is a 27B multimodal model that turns a state and a schema of typed questions into decisions.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":10785,"used_storage":54989721117,"created":"2026-09-30T21:15:35Z","first_seen":"2026-09-30T21:15:35Z","fresh":false,"url":"https://huggingface.co/Cloudflare/clef","likes":2267,"downloads":24097,"finetune":true,"gated":false,"quants":[{"id":"bartowski/Cloudflare_clef-GGUF","kind":"GGUF"},{"id":"ggml-org/Clef-GGUF","kind":"GGUF"},{"id":"mlx-community/clef-4bit","kind":"MLX"},{"id":"mlx-community/clef-8bit","kind":"MLX"},{"id":"mlx-community/clef-OptiQ-4bit","kind":"MLX"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"2267 likes, 24097 downloads","velocity":45.6,"lane":"main","quant_sizes":{"Q4":{"gb":17.7,"file":"Q4_K_M","repo":"bartowski/Cloudflare_clef-GGUF","kind":"GGUF","community":false},"Q8":{"gb":28.8,"file":"Q8_0","repo":"bartowski/Cloudflare_clef-GGUF","kind":"GGUF","community":false},"FP16":{"gb":54.1,"file":"BF16","repo":"bartowski/Cloudflare_clef-GGUF","kind":"GGUF","community":false}},"summary":"Clef is a 27B multimodal model that turns a state and a schema of typed questions into decisions. Dense, 27B params. Context: 256K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/clef","est":{"weights":{"Q4":17.7,"Q8":28.8,"FP16":54.1},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"bartowski/Cloudflare_clef-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"bartowski/Cloudflare_clef-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"bartowski/Cloudflare_clef-GGUF","fmt":"GGUF","community":false}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":20.8,"Q8":31.9,"FP16":57.2},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"Cloudflare/clef-flash","name":"clef-flash","org":"Cloudflare","date":"2026-09-30","license":"apache-2.0","params":9409813744,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":131072,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"Clef-Flash is a 9B multimodal model that turns a state and a schema of typed questions into decisions.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":10813,"used_storage":19083248461,"created":"2026-09-30T21:15:36Z","first_seen":"2026-09-30T21:15:36Z","fresh":false,"url":"https://huggingface.co/Cloudflare/clef-flash","likes":871,"downloads":41461,"finetune":true,"gated":false,"quants":[{"id":"bartowski/Cloudflare_clef-flash-GGUF","kind":"GGUF"},{"id":"ggml-org/Clef-Flash-GGUF","kind":"GGUF"},{"id":"mlx-community/clef-flash-4bit","kind":"MLX"},{"id":"mlx-community/clef-flash-8bit","kind":"MLX"},{"id":"mlx-community/clef-flash-OptiQ-4bit","kind":"MLX"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"871 likes, 41461 downloads","velocity":73.22,"lane":"main","quant_sizes":{"Q4":{"gb":6.0,"file":"Q4_K_M","repo":"bartowski/Cloudflare_clef-flash-GGUF","kind":"GGUF","community":false},"Q8":{"gb":9.7,"file":"Q8_0","repo":"bartowski/Cloudflare_clef-flash-GGUF","kind":"GGUF","community":false},"FP16":{"gb":18.2,"file":"BF16","repo":"bartowski/Cloudflare_clef-flash-GGUF","kind":"GGUF","community":false}},"summary":"Clef-Flash is a 9B multimodal model that turns a state and a schema of typed questions into decisions. Dense, 9.4B params. Context: 256K tokens (config). License: apache-2.0.","ollama":"https://ollama.com/library/clef-flash","est":{"weights":{"Q4":6.0,"Q8":9.7,"FP16":18.2},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"bartowski/Cloudflare_clef-flash-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"bartowski/Cloudflare_clef-flash-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"bartowski/Cloudflare_clef-flash-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":8.1,"Q8":11.8,"FP16":20.3},"fit":{"nv8":null,"gpu12":"Q4","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"Q8","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"Cactus-Compute/whistle","name":"whistle","org":"Cactus-Compute","date":"2026-09-30","license":"apache-2.0","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"automatic-speech-recognition","modality":"audio","summary_what":"Whistle is a log-mel front end and a convolutional stem feeding an audio encoder, read by a Needle-shaped decoder through gated cross attention at every layer.","arch":{"class":"WhistleForSpeechRecognition"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":7050,"used_storage":237538027,"created":"2026-09-30T17:13:26Z","first_seen":"2026-09-30T17:13:26Z","fresh":false,"url":"https://huggingface.co/Cactus-Compute/whistle","likes":507,"downloads":54876,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"507 likes, 54876 downloads","velocity":95.13,"lane":"main","quant_sizes":{},"summary":"Whistle is a log-mel front end and a convolutional stem feeding an audio encoder, read by a Needle-shaped decoder through gated cross attention at every layer. License: apache-2.0.","est":null,"license_class":"permissive"},{"type":"model","id":"KittenML/kitten-tts-2","name":"kitten-tts-2","org":"KittenML","date":"2026-09-30","license":"stellon-labs-community-license","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"text-to-speech","modality":"audio","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":6878,"used_storage":20713390977,"created":"2026-09-30T00:43:12Z","first_seen":"2026-09-30T00:43:12Z","fresh":false,"url":"https://huggingface.co/KittenML/kitten-tts-2","likes":180,"downloads":50188,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"180 likes, 50188 downloads","velocity":84.13,"lane":"main","quant_sizes":{},"summary":"Text to speech model from KittenML (HF pipeline tag; the card has no one-line description). License: stellon-labs-community-license.","est":null,"license_class":"restricted"},{"type":"model","id":"speach1sdef178/MiniMax-H3-X2-Detail-VAE","name":"MiniMax-H3-X2-Detail-VAE","org":"speach1sdef178","date":"2026-09-30","license":"other","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"image-to-video","modality":"video","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":19637,"used_storage":5257252744,"created":"2026-09-30T19:59:27Z","first_seen":"2026-09-30T19:59:27Z","fresh":false,"url":"https://huggingface.co/speach1sdef178/MiniMax-H3-X2-Detail-VAE","likes":103,"downloads":38823,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"103 likes, 38823 downloads","velocity":67.18,"lane":"main","quant_sizes":{},"summary":"Image to video model from speach1sdef178 (HF pipeline tag; the card has no one-line description). License: other.","est":null,"license_class":"restricted"},{"type":"model","id":"facebook/meta-encoder","name":"meta-encoder","org":"facebook","date":"2026-09-30","license":"apache-2.0","params":29776626688,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["BF16"],"pipeline":"feature-extraction","modality":"embedding","summary_what":"","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":9147,"used_storage":59581563633,"created":"2026-09-30T18:00:19Z","first_seen":"2026-09-30T18:00:19Z","fresh":false,"url":"https://huggingface.co/facebook/meta-encoder","likes":28,"downloads":140,"finetune":true,"gated":true,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.29,"lane":"main","quant_sizes":{"Q4":{"gb":19.7,"file":"Q4_K_XL","repo":"webmp3/Sakura-MetaEncoder-30B-Experiment-GGUF","kind":"GGUF","community":true},"FP16":{"gb":59.6,"file":"official BF16 safetensors","repo":"facebook/meta-encoder","kind":"safetensors","community":false}},"summary":"Feature extraction model from facebook (HF pipeline tag; the card has no one-line description). 30B params. License: apache-2.0.","est":{"weights":{"Q4":19.7,"Q8":31.6,"FP16":59.6},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_XL","repo":"webmp3/Sakura-MetaEncoder-30B-Experiment-GGUF","fmt":"GGUF","community":true},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 29.8B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"facebook/meta-encoder","fmt":"safetensors","community":false}},"kv_gb":2.5,"kv_src":"rough","need":{"Q4":23.2,"Q8":35.1,"FP16":63.1},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":null,"strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"real","nv48":"estimate","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"estimate","mac64":"estimate","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"estimate","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"BAAI/AREX-2","name":"AREX-2","org":"BAAI","date":"2026-09-29","license":"apache-2.0","params":27356728560,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"AREX-2 is a 27B-parameter long-horizon agent model from the Beijing Academy of Artificial Intelligence (BAAI).","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":14482,"used_storage":54735776890,"created":"2026-09-29T17:56:21Z","first_seen":"2026-09-29T17:56:21Z","fresh":false,"url":"https://huggingface.co/BAAI/AREX-2","likes":128,"downloads":2423,"finetune":true,"gated":false,"quants":[{"id":"bartowski/BAAI_AREX-2-GGUF","kind":"GGUF"},{"id":"mlx-community/AREX-2-4bit","kind":"MLX"},{"id":"mlx-community/AREX-2-5bit","kind":"MLX"},{"id":"mlx-community/AREX-2-6bit","kind":"MLX"},{"id":"mlx-community/AREX-2-8bit","kind":"MLX"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"128 likes","velocity":4.21,"lane":"main","quant_sizes":{"Q4":{"gb":17.2,"file":"Q4_K_M","repo":"bartowski/BAAI_AREX-2-GGUF","kind":"GGUF","community":false},"Q8":{"gb":28.7,"file":"Q8_0","repo":"bartowski/BAAI_AREX-2-GGUF","kind":"GGUF","community":false},"FP16":{"gb":53.8,"file":"BF16","repo":"bartowski/BAAI_AREX-2-GGUF","kind":"GGUF","community":false}},"summary":"AREX-2 is a 27B-parameter long-horizon agent model from the Beijing Academy of Artificial Intelligence (BAAI). Dense, 27B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":17.2,"Q8":28.7,"FP16":53.8},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"bartowski/BAAI_AREX-2-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"bartowski/BAAI_AREX-2-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"bartowski/BAAI_AREX-2-GGUF","fmt":"GGUF","community":false}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":20.3,"Q8":31.8,"FP16":56.9},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"Q4","nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":"Q4","strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q8","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":"real","strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"microsoft/AesCode-8B","name":"AesCode-8B","org":"microsoft","date":"2026-09-29","license":"apache-2.0","params":8767123696,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":147456,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"","arch":{"class":"Qwen3VLForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":7960,"used_storage":17546162498,"created":"2026-09-29T13:04:44Z","first_seen":"2026-09-29T13:04:44Z","fresh":false,"url":"https://huggingface.co/microsoft/AesCode-8B","likes":78,"downloads":1460,"finetune":true,"gated":false,"quants":[{"id":"bartowski/AesCode-8B-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":2.52,"lane":"main","quant_sizes":{"Q4":{"gb":5.2,"file":"Q4_K_M","repo":"bartowski/AesCode-8B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":8.7,"file":"Q8_0","repo":"bartowski/AesCode-8B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":16.4,"file":"BF16","repo":"bartowski/AesCode-8B-GGUF","kind":"GGUF","community":false}},"summary":"Image text to text model from microsoft (HF pipeline tag; the card has no one-line description). Dense, 8.8B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":5.2,"Q8":8.7,"FP16":16.4},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"bartowski/AesCode-8B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"bartowski/AesCode-8B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"bartowski/AesCode-8B-GGUF","fmt":"GGUF","community":false}},"kv_gb":1.2,"kv_src":"config","need":{"Q4":7.4,"Q8":10.9,"FP16":18.6},"fit":{"nv8":null,"gpu12":"Q8","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"Q8","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"microsoft/AesCode-32B","name":"AesCode-32B","org":"microsoft","date":"2026-09-29","license":"apache-2.0","params":33357390064,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":262144,"dtypes":["BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"","arch":{"class":"Qwen3VLForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":8482,"used_storage":66726735794,"created":"2026-09-29T13:05:36Z","first_seen":"2026-09-29T13:05:36Z","fresh":false,"url":"https://huggingface.co/microsoft/AesCode-32B","likes":55,"downloads":1358,"finetune":true,"gated":false,"quants":[{"id":"bartowski/AesCode-32B-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":2.32,"lane":"main","quant_sizes":{"Q4":{"gb":20.5,"file":"Q4_K_M","repo":"bartowski/AesCode-32B-GGUF","kind":"GGUF","community":false},"Q8":{"gb":34.8,"file":"Q8_0","repo":"bartowski/AesCode-32B-GGUF","kind":"GGUF","community":false},"FP16":{"gb":65.5,"file":"BF16","repo":"bartowski/AesCode-32B-GGUF","kind":"GGUF","community":false}},"summary":"Image text to text model from microsoft (HF pipeline tag; the card has no one-line description). Dense, 33B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":20.5,"Q8":34.8,"FP16":65.5},"weights_src":{"Q4":{"kind":"real","file":"Q4_K_M","repo":"bartowski/AesCode-32B-GGUF","fmt":"GGUF","community":false},"Q8":{"kind":"real","file":"Q8_0","repo":"bartowski/AesCode-32B-GGUF","fmt":"GGUF","community":false},"FP16":{"kind":"real","file":"BF16","repo":"bartowski/AesCode-32B-GGUF","fmt":"GGUF","community":false}},"kv_gb":2.1,"kv_src":"config","need":{"Q4":23.6,"Q8":37.9,"FP16":68.6},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":null,"strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q4","mac64":"Q8","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"internlm/AdvancedMathBench-AutoVerifier","name":"AdvancedMathBench-AutoVerifier","org":"internlm","date":"2026-09-29","license":"","params":35964405616,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":81920,"dtypes":["BF16","F32"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"AutoVerifier is a learned grader, not a formal proof checker, and can make errors.","arch":{"class":"Qwen3_5MoeForConditionalGeneration","type":"MoE","experts":256,"experts_active":8},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":3353,"used_storage":72971585006,"created":"2026-09-29T06:53:37Z","first_seen":"2026-09-29T06:53:37Z","fresh":false,"url":"https://huggingface.co/internlm/AdvancedMathBench-AutoVerifier","likes":1,"downloads":312,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":0.51,"lane":"main","quant_sizes":{"FP16":{"gb":73.0,"file":"official BF16 safetensors","repo":"internlm/AdvancedMathBench-AutoVerifier","kind":"safetensors","community":false}},"summary":"AutoVerifier is a learned grader, not a formal proof checker, and can make errors. Mixture-of-experts, 36B total (256 experts, 8 per token). Context: 256K tokens (config).","est":{"weights":{"Q4":21.6,"Q8":38.2,"FP16":73.0},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 36.0B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 36.0B params"},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"internlm/AdvancedMathBench-AutoVerifier","fmt":"safetensors","community":false}},"kv_gb":0.7,"kv_src":"config","need":{"Q4":23.3,"Q8":39.9,"FP16":74.7},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"Q4","nv48":"Q8","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":null,"amd24":null,"strix":"FP16","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"Q4","mac48":"Q4","mac64":"Q8","mac96":"Q8","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"Q4","cpu64":"Q8","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":"estimate","nv48":"estimate","nv96":"real","spark1":"real","spark2":"real","amd16":null,"amd24":null,"strix":"real","gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"estimate","cpu64":"estimate","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"unknown"},{"type":"model","id":"NaiveAI/Naive-N0.5-Flash","name":"Naive-N0.5-Flash","org":"NaiveAI","date":"2026-09-27","license":"mit","params":null,"params_src":null,"active":15500000000.0,"ctx":1048576,"kv_bytes_tok":147456,"dtypes":[],"pipeline":"text-generation","modality":"text","summary_what":"Naive-N0.5-Flash is an open-weight 309B MoE model with 15.5B active parameters, built for coding and AI R&D.","arch":{"class":"NaiveN05FlashForCausalLM","type":"MoE","experts":256,"experts_active":8,"active":15500000000.0,"active_src":"model card"},"claims":["SWA uses a 128-token window, while DSA selects the top 2,048 tokens for backbone attention."],"has_safetensors":true,"has_weights":true,"card_len":11377,"used_storage":617823283433,"created":"2026-09-27T14:14:32Z","first_seen":"2026-09-27T14:14:32Z","fresh":false,"url":"https://huggingface.co/NaiveAI/Naive-N0.5-Flash","likes":292,"downloads":4481,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"292 likes","velocity":7.26,"lane":"main","quant_sizes":{"Q8":{"gb":0.7,"file":"Q8_0","repo":"Baekpica/Naive-N0.5-Flash-Mixed-Quant-GGUF","kind":"GGUF","community":true}},"summary":"Naive-N0.5-Flash is an open-weight 309B MoE model with 15.5B active parameters, built for coding and AI R&D. Mixture-of-experts, 16B active (256 experts, 8 per token). Context: 1M tokens (config). License: mit. Vendor-reported: \"SWA uses a 128-token window, while DSA selects the top 2,048 tokens for backbone attention.\"","est":{"weights":{"Q8":0.7},"weights_src":{"Q8":{"kind":"real","file":"Q8_0","repo":"Baekpica/Naive-N0.5-Flash-Mixed-Quant-GGUF","fmt":"GGUF","community":true}},"kv_gb":1.2,"kv_src":"config","need":{"Q8":2.9},"fit":{"nv8":"Q8","gpu12":"Q8","gpu16":"Q8","nv24":"Q8","nv32":"Q8","nv48":"Q8","nv96":"Q8","spark1":"Q8","spark2":"Q8","amd16":"Q8","amd24":"Q8","strix":"Q8","gpu24":"Q8","gpu32":"Q8","mac16":"Q8","mac24":"Q8","mac36":"Q8","mac48":"Q8","mac64":"Q8","mac96":"Q8","mac128":"Q8","mac192":"Q8","mac256":"Q8","mac512":"Q8","cpu32":"Q8","cpu64":"Q8","cpu128":"Q8","cpu256":"Q8"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"real quant files"},"license_class":"permissive"},{"type":"model","id":"shhivv/taiga-s1","name":"taiga-s1","org":"shhivv","date":"2026-09-27","license":"mit","params":1228163,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"other","modality":"other","summary_what":"Taiga-S1 is the fast \"System 1\" layer for a CAD agent.","arch":{},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":4930,"used_storage":6777700,"created":"2026-09-27T13:34:07Z","first_seen":"2026-09-27T13:34:07Z","fresh":false,"url":"https://huggingface.co/shhivv/taiga-s1","likes":157,"downloads":927,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"157 likes","velocity":1.65,"lane":"main","quant_sizes":{},"summary":"Taiga-S1 is the fast \"System 1\" layer for a CAD agent. 1M params. License: mit.","est":{"weights":{"Q4":0.0,"Q8":0.0,"FP16":0.0},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.0B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 0.0B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 0.0B params"}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":1.5,"Q8":1.5,"FP16":1.5},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"estimate","gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"Aratako/Irodori-TTS-v4-Large","name":"Irodori-TTS-v4-Large","org":"Aratako","date":"2026-09-27","license":"gemma","params":3288179489,"params_src":"safetensors","active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":["F32"],"pipeline":"text-to-speech","modality":"audio","summary_what":"Irodori-TTS-v4-Large is a Japanese Text-to-Speech model based on a Rectified Flow Diffusion Transformer (RF-DiT) architecture.","arch":{},"claims":["v4-Large achieves higher mean CAM++ cosine similarity at all four reference lengths, although 120-second top-1 accuracy is slightly lower."],"has_safetensors":true,"has_weights":true,"card_len":14313,"used_storage":13186277735,"created":"2026-09-27T12:21:00Z","first_seen":"2026-09-27T12:21:00Z","fresh":false,"url":"https://huggingface.co/Aratako/Irodori-TTS-v4-Large","likes":78,"downloads":0,"finetune":false,"gated":false,"quants":[{"id":"mlx-community/Irodori-TTS-v4-Large-8bit","kind":"MLX"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"78 likes","velocity":0.12,"lane":"main","quant_sizes":{"Q8":{"gb":3.9,"file":"Q8_0","repo":"sakasegawa/Irodori-TTS-v4-Large-GGUF","kind":"GGUF","community":true},"FP16":{"gb":7.0,"file":"F16","repo":"sakasegawa/Irodori-TTS-v4-Large-GGUF","kind":"GGUF","community":true}},"summary":"Irodori-TTS-v4-Large is a Japanese Text-to-Speech model based on a Rectified Flow Diffusion Transformer (RF-DiT) architecture. 3.3B params. License: gemma. Vendor-reported: \"v4-Large achieves higher mean CAM++ cosine similarity at all four reference lengths, although 120-second top-1 accuracy is slightly lower.\"","est":{"weights":{"Q4":2.0,"Q8":3.9,"FP16":7.0},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 3.3B params"},"Q8":{"kind":"real","file":"Q8_0","repo":"sakasegawa/Irodori-TTS-v4-Large-GGUF","fmt":"GGUF","community":true},"FP16":{"kind":"real","file":"F16","repo":"sakasegawa/Irodori-TTS-v4-Large-GGUF","fmt":"GGUF","community":true}},"kv_gb":0.5,"kv_src":"rough","need":{"Q4":3.5,"Q8":5.4,"FP16":8.5},"fit":{"nv8":"Q8","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"restricted"},{"type":"model","id":"XiaomiMiMo/MiMo-V2.6-Flash-MOPD","name":"MiMo-V2.6-Flash-MOPD","org":"XiaomiMiMo","date":"2026-09-27","license":"mit","params":310756322688,"params_src":"safetensors","active":15000000000.0,"ctx":1048576,"kv_bytes_tok":147456,"dtypes":["F32","BF16","F8_E4M3","U8"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"MiMoV2ForCausalLM","type":"MoE","experts":256,"experts_active":8,"active":15000000000.0,"active_src":"model card"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":8797,"used_storage":177753496517,"created":"2026-09-27T04:01:00Z","first_seen":"2026-09-27T04:01:00Z","fresh":false,"url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-MOPD","likes":68,"downloads":12418,"finetune":false,"gated":false,"quants":[{"id":"ggml-org/MiMo-V2.6-Flash-MOPD-GGUF","kind":"GGUF"}],"x_mentions":[{"url":"https://x.com/Chris_0716/status/2112742020175610293","user":"Chris_0716","likes":3},{"url":"https://x.com/Chris_0716/status/2112738287224045767","user":"Chris_0716","likes":0}],"x_likes":3,"x_posts":2,"signal":"trusted org","velocity":38.71,"lane":"main","quant_sizes":{},"summary":"Text generation model from XiaomiMiMo (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 311B total, 15B active (256 experts, 8 per token). Context: 1M tokens (config). License: mit.","est":{"weights":{"Q4":186.5,"Q8":330.2,"FP16":621.5},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 310.8B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 310.8B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 310.8B params"}},"kv_gb":1.2,"kv_src":"config","need":{"Q4":188.7,"Q8":332.4,"FP16":623.7},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":"Q4","amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":"Q4","mac512":"Q8","cpu32":null,"cpu64":null,"cpu128":null,"cpu256":"Q4"},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":"estimate","amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":"estimate","mac512":"estimate","cpu32":null,"cpu64":null,"cpu128":null,"cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"XiaomiMiMo/MiMo-V2.6-Pro-MOPD","name":"MiMo-V2.6-Pro-MOPD","org":"XiaomiMiMo","date":"2026-09-27","license":"mit","params":1024216603392,"params_src":"safetensors","active":42000000000.0,"ctx":1048576,"kv_bytes_tok":430080,"dtypes":["BF16","F32","F8_E4M3","U8"],"pipeline":"text-generation","modality":"text","summary_what":"","arch":{"class":"MiMoV2ForCausalLM","type":"MoE","experts":384,"experts_active":8,"active":42000000000.0,"active_src":"model card"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":8958,"used_storage":573484895920,"created":"2026-09-27T03:58:00Z","first_seen":"2026-09-27T03:58:00Z","fresh":false,"url":"https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Pro-MOPD","likes":44,"downloads":9551,"finetune":false,"gated":false,"quants":[{"id":"ggml-org/MiMo-V2.6-Pro-MOPD-GGUF","kind":"GGUF"}],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":14.38,"lane":"main","quant_sizes":{},"summary":"Text generation model from XiaomiMiMo (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 1024B total, 42B active (384 experts, 8 per token). Context: 1M tokens (config). License: mit.","est":{"weights":{"Q4":614.5,"Q8":1088.2,"FP16":2048.4},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 1024.2B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 1024.2B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 1024.2B params"}},"kv_gb":3.5,"kv_src":"config","need":{"Q4":619.0,"Q8":1092.7,"FP16":2052.9},"fit":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":null,"amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":null,"mac512":null,"cpu32":null,"cpu64":null,"cpu128":null,"cpu256":null},"fit_src":{"nv8":null,"gpu12":null,"gpu16":null,"nv24":null,"nv32":null,"nv48":null,"nv96":null,"spark1":null,"spark2":null,"amd16":null,"amd24":null,"strix":null,"gpu24":null,"gpu32":null,"mac16":null,"mac24":null,"mac36":null,"mac48":null,"mac64":null,"mac96":null,"mac128":null,"mac192":null,"mac256":null,"mac512":null,"cpu32":null,"cpu64":null,"cpu128":null,"cpu256":null},"basis":"estimate from params"},"license_class":"permissive"},{"type":"model","id":"internlm/Intern-Decision-4B","name":"Intern-Decision-4B","org":"internlm","date":"2026-09-26","license":"apache-2.0","params":4539265536,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":131072,"dtypes":["F32","BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"Intern-Decision-4B is a multimodal structured decision model fine-tuned from Qwen3.5-4B.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":10021,"used_storage":9098616266,"created":"2026-09-26T05:36:37Z","first_seen":"2026-09-26T05:36:37Z","fresh":false,"url":"https://huggingface.co/internlm/Intern-Decision-4B","likes":90,"downloads":5220,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":7.7,"lane":"main","quant_sizes":{"Q8":{"gb":4.5,"file":"Q8_0","repo":"bombdefuser-124/Intern-Decision-4B-GGUF","kind":"GGUF","community":true},"FP16":{"gb":9.1,"file":"official BF16 safetensors","repo":"internlm/Intern-Decision-4B","kind":"safetensors","community":false}},"summary":"Intern-Decision-4B is a multimodal structured decision model fine-tuned from Qwen3.5-4B. Dense, 4.5B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":2.7,"Q8":4.5,"FP16":9.1},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 4.5B params"},"Q8":{"kind":"real","file":"Q8_0","repo":"bombdefuser-124/Intern-Decision-4B-GGUF","fmt":"GGUF","community":true},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"internlm/Intern-Decision-4B","fmt":"safetensors","community":false}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":4.8,"Q8":6.6,"FP16":11.2},"fit":{"nv8":"Q8","gpu12":"Q8","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"internlm/Intern-Decision-0.8B","name":"Intern-Decision-0.8B","org":"internlm","date":"2026-09-26","license":"apache-2.0","params":852985920,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":49152,"dtypes":["F32","BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"Intern-Decision-0.8B is a multimodal structured decision model fine-tuned from Qwen3.5-0.8B.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":9036,"used_storage":1726024266,"created":"2026-09-26T05:35:57Z","first_seen":"2026-09-26T05:35:57Z","fresh":false,"url":"https://huggingface.co/internlm/Intern-Decision-0.8B","likes":37,"downloads":3672,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":5.38,"lane":"main","quant_sizes":{"Q8":{"gb":0.8,"file":"Q8_0","repo":"bombdefuser-124/Intern-Decision-0.8B-GGUF","kind":"GGUF","community":true},"FP16":{"gb":1.7,"file":"official BF16 safetensors","repo":"internlm/Intern-Decision-0.8B","kind":"safetensors","community":false}},"summary":"Intern-Decision-0.8B is a multimodal structured decision model fine-tuned from Qwen3.5-0.8B. Dense, 853M params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":0.5,"Q8":0.8,"FP16":1.7},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 0.9B params"},"Q8":{"kind":"real","file":"Q8_0","repo":"bombdefuser-124/Intern-Decision-0.8B-GGUF","fmt":"GGUF","community":true},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"internlm/Intern-Decision-0.8B","fmt":"safetensors","community":false}},"kv_gb":0.4,"kv_src":"config","need":{"Q4":1.9,"Q8":2.2,"FP16":3.1},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"internlm/Intern-Decision-2B","name":"Intern-Decision-2B","org":"internlm","date":"2026-09-26","license":"apache-2.0","params":2213241664,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":49152,"dtypes":["F32","BF16"],"pipeline":"image-text-to-text","modality":"vision","summary_what":"Intern-Decision-2B is a multimodal structured decision model fine-tuned from Qwen3.5-2B.","arch":{"class":"Qwen3_5ForConditionalGeneration","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":9028,"used_storage":4446552362,"created":"2026-09-26T05:36:19Z","first_seen":"2026-09-26T05:36:19Z","fresh":false,"url":"https://huggingface.co/internlm/Intern-Decision-2B","likes":18,"downloads":2591,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":3.78,"lane":"main","quant_sizes":{"Q8":{"gb":2.0,"file":"Q8_0","repo":"bombdefuser-124/Intern-Decision-2B-GGUF","kind":"GGUF","community":true},"FP16":{"gb":4.4,"file":"official BF16 safetensors","repo":"internlm/Intern-Decision-2B","kind":"safetensors","community":false}},"summary":"Intern-Decision-2B is a multimodal structured decision model fine-tuned from Qwen3.5-2B. Dense, 2.2B params. Context: 256K tokens (config). License: apache-2.0.","est":{"weights":{"Q4":1.3,"Q8":2.0,"FP16":4.4},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 2.2B params"},"Q8":{"kind":"real","file":"Q8_0","repo":"bombdefuser-124/Intern-Decision-2B-GGUF","fmt":"GGUF","community":true},"FP16":{"kind":"real","file":"official BF16 safetensors","repo":"internlm/Intern-Decision-2B","fmt":"safetensors","community":false}},"kv_gb":0.4,"kv_src":"config","need":{"Q4":2.7,"Q8":3.4,"FP16":5.8},"fit":{"nv8":"FP16","gpu12":"FP16","gpu16":"FP16","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"FP16","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"FP16","mac24":"FP16","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":"real","gpu12":"real","gpu16":"real","nv24":"real","nv32":"real","nv48":"real","nv96":"real","spark1":"real","spark2":"real","amd16":"real","amd24":"real","strix":"real","gpu24":"real","gpu32":"real","mac16":"real","mac24":"real","mac36":"real","mac48":"real","mac64":"real","mac96":"real","mac128":"real","mac192":"real","mac256":"real","mac512":"real","cpu32":"real","cpu64":"real","cpu128":"real","cpu256":"real"},"basis":"mixed: real files + estimate"},"license_class":"permissive"},{"type":"model","id":"Veda-Sparse/Minimax-H3-T2VA-Veda-8NFE-600Step-Preview","name":"Minimax-H3-T2VA-Veda-8NFE-600Step-Preview","org":"Veda-Sparse","date":"2026-09-25","license":"minimax-h3-community","params":null,"params_src":null,"active":null,"ctx":null,"kv_bytes_tok":null,"dtypes":[],"pipeline":"text-to-video","modality":"video","summary_what":"Veda is a learned sparse-attention method for video diffusion models.","arch":{"type":"dense"},"claims":["Visual and audio quality on par with full attention while skipping 90% of it."],"has_safetensors":true,"has_weights":true,"card_len":7562,"used_storage":299933306,"created":"2026-09-25T10:56:07Z","first_seen":"2026-09-25T10:56:07Z","fresh":false,"url":"https://huggingface.co/Veda-Sparse/Minimax-H3-T2VA-Veda-8NFE-600Step-Preview","likes":115,"downloads":27665,"finetune":true,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"115 likes, 27665 downloads","velocity":39.21,"lane":"main","quant_sizes":{},"summary":"Veda is a learned sparse-attention method for video diffusion models. Dense. License: minimax-h3-community. Vendor-reported: \"Visual and audio quality on par with full attention while skipping 90% of it.\"","est":null,"license_class":"restricted"},{"type":"model","id":"perplexity-ai/pplx-embed-v2-context-9b-preview","name":"pplx-embed-v2-context-9b-preview","org":"perplexity-ai","date":"2026-09-25","license":"mit","params":8401083632,"params_src":"safetensors","active":null,"ctx":262144,"kv_bytes_tok":131072,"dtypes":["F32"],"pipeline":"feature-extraction","modality":"embedding","summary_what":"","arch":{"class":"PplxContextualModel","type":"dense"},"claims":[],"has_safetensors":true,"has_weights":true,"card_len":3881,"used_storage":67228835880,"created":"2026-09-25T17:22:14Z","first_seen":"2026-09-25T17:22:14Z","fresh":false,"url":"https://huggingface.co/perplexity-ai/pplx-embed-v2-context-9b-preview","likes":74,"downloads":1861,"finetune":false,"gated":false,"quants":[],"x_mentions":[],"x_likes":0,"x_posts":0,"signal":"trusted org","velocity":2.76,"lane":"main","quant_sizes":{},"summary":"Feature extraction model from perplexity-ai (HF pipeline tag; the card has no one-line description). Dense, 8.4B params. Context: 256K tokens (config). License: mit.","est":{"weights":{"Q4":5.0,"Q8":8.9,"FP16":16.8},"weights_src":{"Q4":{"kind":"estimate","file":"~4.8 bits/weight x 8.4B params"},"Q8":{"kind":"estimate","file":"~8.5 bits/weight x 8.4B params"},"FP16":{"kind":"estimate","file":"~16.0 bits/weight x 8.4B params"}},"kv_gb":1.1,"kv_src":"config","need":{"Q4":7.1,"Q8":11.0,"FP16":18.9},"fit":{"nv8":null,"gpu12":"Q8","gpu16":"Q8","nv24":"FP16","nv32":"FP16","nv48":"FP16","nv96":"FP16","spark1":"FP16","spark2":"FP16","amd16":"Q8","amd24":"FP16","strix":"FP16","gpu24":"FP16","gpu32":"FP16","mac16":"Q8","mac24":"Q8","mac36":"FP16","mac48":"FP16","mac64":"FP16","mac96":"FP16","mac128":"FP16","mac192":"FP16","mac256":"FP16","mac512":"FP16","cpu32":"FP16","cpu64":"FP16","cpu128":"FP16","cpu256":"FP16"},"fit_src":{"nv8":null,"gpu12":"estimate","gpu16":"estimate","nv24":"estimate","nv32":"estimate","nv48":"estimate","nv96":"estimate","spark1":"estimate","spark2":"estimate","amd16":"estimate","amd24":"estimate","strix":"estimate","gpu24":"estimate","gpu32":"estimate","mac16":"estimate","mac24":"estimate","mac36":"estimate","mac48":"estimate","mac64":"estimate","mac96":"estimate","mac128":"estimate","mac192":"estimate","mac256":"estimate","mac512":"estimate","cpu32":"estimate","cpu64":"estimate","cpu128":"estimate","cpu256":"estimate"},"basis":"estimate from params"},"license_class":"permissive"}],"filtered_models":[{"id":"gradium/phonon-40m-multilingual","org":"gradium","date":"2026-10-19","likes":47,"downloads":2331,"x_likes":0,"finetune":false,"url":"https://huggingface.co/gradium/phonon-40m-multilingual","filter_reason":"aged out below threshold"},{"id":"hazyresearch/ncm-checkpoints","org":"hazyresearch","date":"2026-10-13","likes":44,"downloads":0,"x_likes":0,"finetune":false,"url":"https://huggingface.co/hazyresearch/ncm-checkpoints","filter_reason":"aged out below threshold"},{"id":"Kijai/MOSS-image_experimental","org":"Kijai","date":"2026-10-16","likes":38,"downloads":0,"x_likes":0,"finetune":false,"url":"https://huggingface.co/Kijai/MOSS-image_experimental","filter_reason":"aged out below threshold"},{"id":"XHToken/Spark-X2.6-4B-Base","org":"XHToken","date":"2026-10-19","likes":27,"downloads":263,"x_likes":0,"finetune":false,"url":"https://huggingface.co/XHToken/Spark-X2.6-4B-Base","filter_reason":"aged out below threshold"},{"id":"VAST-AI/EdenLive","org":"VAST-AI","date":"2026-10-12","likes":21,"downloads":297,"x_likes":0,"finetune":false,"url":"https://huggingface.co/VAST-AI/EdenLive","filter_reason":"aged out below threshold"},{"id":"Blockway/Agens-Volundr-32B-v1","org":"Blockway","date":"2026-10-17","likes":15,"downloads":305,"x_likes":0,"finetune":false,"url":"https://huggingface.co/Blockway/Agens-Volundr-32B-v1","filter_reason":"aged out below threshold"},{"id":"gradium/phonon-voices","org":"gradium","date":"2026-10-19","likes":15,"downloads":0,"x_likes":0,"finetune":false,"url":"https://huggingface.co/gradium/phonon-voices","filter_reason":"aged out below threshold"},{"id":"gradium/phonon-40m-fr","org":"gradium","date":"2026-10-19","likes":17,"downloads":2058,"x_likes":0,"finetune":false,"url":"https://huggingface.co/gradium/phonon-40m-fr","filter_reason":"aged out below threshold"},{"id":"gradium/phonon-40m-es","org":"gradium","date":"2026-10-19","likes":15,"downloads":1961,"x_likes":0,"finetune":false,"url":"https://huggingface.co/gradium/phonon-40m-es","filter_reason":"aged out below threshold"},{"id":"gradium/phonon-40m-de","org":"gradium","date":"2026-10-19","likes":13,"downloads":4541,"x_likes":0,"finetune":false,"url":"https://huggingface.co/gradium/phonon-40m-de","filter_reason":"aged out below threshold"},{"id":"Teriminal-RSI/Terminal-RSI-27B","org":"Teriminal-RSI","date":"2026-10-21","likes":12,"downloads":621,"x_likes":0,"finetune":false,"url":"https://huggingface.co/Teriminal-RSI/Terminal-RSI-27B","filter_reason":"aged out below threshold"},{"id":"gradium/phonon-40m-pt","org":"gradium","date":"2026-10-19","likes":14,"downloads":733,"x_likes":0,"finetune":false,"url":"https://huggingface.co/gradium/phonon-40m-pt","filter_reason":"aged out below threshold"},{"id":"startlux-models/StartLux-Decision-v1.1-27B-Instant","org":"startlux-models","date":"2026-10-11","likes":12,"downloads":4405,"x_likes":0,"finetune":false,"url":"https://huggingface.co/startlux-models/StartLux-Decision-v1.1-27B-Instant","filter_reason":"aged out below threshold"},{"id":"Edisoneh/InfiniSplat-v2","org":"Edisoneh","date":"2026-10-11","likes":10,"downloads":1043,"x_likes":0,"finetune":false,"url":"https://huggingface.co/Edisoneh/InfiniSplat-v2","filter_reason":"aged out below threshold"},{"id":"caikybaldo999/zyi-256-flash","org":"caikybaldo999","date":"2026-10-21","likes":3,"downloads":101,"x_likes":0,"finetune":false,"url":"https://huggingface.co/caikybaldo999/zyi-256-flash","filter_reason":"aged out below threshold"},{"id":"fyxfelixfu/consistentDMD","org":"fyxfelixfu","date":"2026-10-18","likes":2,"downloads":0,"x_likes":0,"finetune":false,"url":"https://huggingface.co/fyxfelixfu/consistentDMD","filter_reason":"aged out below threshold"},{"id":"kaupane/Inko","org":"kaupane","date":"2026-10-16","likes":4,"downloads":0,"x_likes":0,"finetune":false,"url":"https://huggingface.co/kaupane/Inko","filter_reason":"aged out below threshold"},{"id":"public-release-bp/BudgetPix","org":"public-release-bp","date":"2026-10-15","likes":5,"downloads":0,"x_likes":0,"finetune":false,"url":"https://huggingface.co/public-release-bp/BudgetPix","filter_reason":"aged out below threshold"},{"id":"MithrilMan/teacher-in-the-loop","org":"MithrilMan","date":"2026-10-15","likes":4,"downloads":0,"x_likes":0,"finetune":false,"url":"https://huggingface.co/MithrilMan/teacher-in-the-loop","filter_reason":"aged out below threshold"},{"id":"TencentARC/Looking-Ahead","org":"TencentARC","date":"2026-10-21","likes":4,"downloads":0,"x_likes":0,"finetune":false,"url":"https://huggingface.co/TencentARC/Looking-Ahead","filter_reason":"aged out below threshold"},{"id":"robertbschuette/Cobijada_Minimax-H3_Hybrid_Pruned_ComfyUI","org":"robertbschuette","date":"2026-10-19","likes":2,"downloads":2328,"x_likes":0,"finetune":false,"url":"https://huggingface.co/robertbschuette/Cobijada_Minimax-H3_Hybrid_Pruned_ComfyUI","filter_reason":"aged out below threshold"},{"id":"Arko007/Zenyx-100M-Base","org":"Arko007","date":"2026-10-17","likes":2,"downloads":0,"x_likes":0,"finetune":false,"url":"https://huggingface.co/Arko007/Zenyx-100M-Base","filter_reason":"aged out below threshold"},{"id":"HuggingFriends/RoboJudge-9B","org":"HuggingFriends","date":"2026-09-25","likes":2,"downloads":62,"x_likes":0,"finetune":false,"url":"https://huggingface.co/HuggingFriends/RoboJudge-9B","filter_reason":"aged out below threshold"}],"harnesses":[{"type":"harness","id":"ggml-org/llama.cpp","name":"llama.cpp","org":"ggml-org","date":"2026-10-24","version":"b11869","prerelease":true,"releases_30d":30,"recent_tags":["b11869","b11868","b11867","b11866","b11865","b11861","b11860","b11858"],"summary":"jinja : fix select and selectattr argument handling (#31059)","url":"https://github.com/ggml-org/llama.cpp/releases/tag/b11869","repo_url":"https://github.com/ggml-org/llama.cpp","modality":"","license":""},{"type":"harness","id":"ollama/ollama","name":"Ollama","org":"ollama","date":"2026-10-24","version":"v0.40.11","prerelease":false,"releases_30d":14,"recent_tags":["v0.40.11","v0.40.10","v0.40.9","v0.40.8","v0.40.7","v0.40.6","v0.40.5","v0.40.4"],"summary":"Ollama can now generate speech on macOS.","url":"https://github.com/ollama/ollama/releases/tag/v0.40.11","repo_url":"https://github.com/ollama/ollama","modality":"","license":""},{"type":"harness","id":"ashhart/TensorFold","name":"TensorFold","org":"ashhart","date":"2026-10-24","version":"v1.1.4","prerelease":false,"releases_30d":30,"recent_tags":["v1.1.4","v1.1.3","v1.0.0-beta.9","v1.1.2","v1.1.1","v1.1.0","v1.0.0-beta.8","v1.0.12"],"summary":"1.1.4 makes Qwen3.8-27B's cold prompts on CUDA up to five times faster, brings Gemma 4 12B and 31B to the native CUDA engine with their drafters, lets several Flash Next streams verify their drafts in one pass on CUDA, and adds idle unload,","url":"https://github.com/ashhart/TensorFold/releases/tag/v1.1.4","repo_url":"https://github.com/ashhart/TensorFold","modality":"","license":""},{"type":"harness","id":"NousResearch/hermes-agent","name":"Hermes Agent","org":"NousResearch","date":"2026-10-23","version":"v0.22.5","prerelease":false,"releases_30d":8,"recent_tags":["v0.22.5","v0.22.4","v0.22.3","v0.22.2","v0.22.1","v0.22.0","v0.21.7","v0.21.6"],"summary":"Also in the window, undocumented here on purpose: Nous sign-in reworked (every sign-in is the plain device login, the free-tier guest hands over to your account, the result shows who signed in, connectors offer sign-in instead of opening a ","url":"https://github.com/NousResearch/hermes-agent/releases/tag/v0.22.5","repo_url":"https://github.com/NousResearch/hermes-agent","modality":"","license":""},{"type":"harness","id":"unslothai/unsloth","name":"Unsloth","org":"unslothai","date":"2026-10-23","version":"v0.1.910-beta","prerelease":false,"releases_30d":11,"recent_tags":["v0.1.910-beta","v0.1.909-beta","v0.1.908-beta","v0.1.907-beta","v0.1.906-beta","v0.1.905-beta","v0.1.904-beta","v0.1.903-beta"],"summary":"Recursive's Flint, MOSS-Image and NVIDIA's Nemotron Bolt 4B are here!","url":"https://github.com/unslothai/unsloth/releases/tag/v0.1.910-beta","repo_url":"https://github.com/unslothai/unsloth","modality":"","license":""},{"type":"harness","id":"anomalyco/opencode","name":"OpenCode","org":"anomalyco","date":"2026-10-22","version":"v1.19.4","prerelease":false,"releases_30d":8,"recent_tags":["v1.19.4","v1.19.3","v1.19.2","v1.19.1","v1.19.0","v1.18.35","v1.18.34","v1.18.33"],"summary":"","url":"https://github.com/anomalyco/opencode/releases/tag/v1.19.4","repo_url":"https://github.com/anomalyco/opencode","modality":"","license":"","advisory":{"text":"Security incident Oct 17, 2026: rotate provider keys saved in OpenCode Console","url":"https://ctxsec.com/items/opencode-console-provider-key-breach-2026-10","date":"2026-10-17"}},{"type":"harness","id":"NVIDIA/TensorRT-LLM","name":"TensorRT-LLM","org":"NVIDIA","date":"2026-10-22","version":"v1.4.0rc1","prerelease":true,"releases_30d":3,"recent_tags":["v1.4.0rc1","v1.4.0rc0","v1.3.0rc29"],"summary":"","url":"https://github.com/NVIDIA/TensorRT-LLM/releases/tag/v1.4.0rc1","repo_url":"https://github.com/NVIDIA/TensorRT-LLM","modality":"","license":""},{"type":"harness","id":"turboderp-org/exllamav3","name":"ExLlamaV3","org":"turboderp-org","date":"2026-10-20","version":"v1.6.4","prerelease":false,"releases_30d":8,"recent_tags":["v1.6.4","v1.6.3","v1.6.2","v1.6.1","v1.6.0","v1.5.4","v1.5.3","v1.5.2"],"summary":"","url":"https://github.com/turboderp-org/exllamav3/releases/tag/v1.6.4","repo_url":"https://github.com/turboderp-org/exllamav3","modality":"","license":""},{"type":"harness","id":"sgl-project/sglang","name":"SGLang","org":"sgl-project","date":"2026-10-19","version":"v0.5.22","prerelease":false,"releases_30d":2,"recent_tags":["v0.5.22","v0.5.21"],"summary":"SGLang now requires Python 3.11 or newer.","url":"https://github.com/sgl-project/sglang/releases/tag/v0.5.22","repo_url":"https://github.com/sgl-project/sglang","modality":"","license":""},{"type":"harness","id":"Comfy-Org/ComfyUI","name":"ComfyUI","org":"Comfy-Org","date":"2026-10-19","version":"v0.42.1","prerelease":false,"releases_30d":6,"recent_tags":["v0.42.1","v0.42.0","v0.41.0","v0.40.0","v0.39.0","v0.38.0"],"summary":"Fix some crashing issues a few users experienced when running the MOSS image template workflow.","url":"https://github.com/Comfy-Org/ComfyUI/releases/tag/v0.42.1","repo_url":"https://github.com/Comfy-Org/ComfyUI","modality":"","license":""},{"type":"harness","id":"vllm-project/vllm","name":"vLLM","org":"vllm-project","date":"2026-10-18","version":"v0.32.0","prerelease":false,"releases_30d":2,"recent_tags":["v0.32.0","v0.31.0"],"summary":"This release features 814 commits from 363 contributors (121 new)!","url":"https://github.com/vllm-project/vllm/releases/tag/v0.32.0","repo_url":"https://github.com/vllm-project/vllm","modality":"","license":""},{"type":"harness","id":"ml-explore/mlx-lm","name":"MLX-LM","org":"ml-explore","date":"2026-10-16","version":"v0.32.2","prerelease":false,"releases_30d":3,"recent_tags":["v0.32.2","v0.32.1","v0.32.0"],"summary":"","url":"https://github.com/ml-explore/mlx-lm/releases/tag/v0.32.2","repo_url":"https://github.com/ml-explore/mlx-lm","modality":"","license":""},{"type":"harness","id":"open-webui/open-webui","name":"Open WebUI","org":"open-webui","date":"2026-10-16","version":"v0.12.3","prerelease":false,"releases_30d":4,"recent_tags":["v0.12.3","v0.12.2","v0.12.1","v0.12.0"],"summary":"","url":"https://github.com/open-webui/open-webui/releases/tag/v0.12.3","repo_url":"https://github.com/open-webui/open-webui","modality":"","license":""},{"type":"harness","id":"intel/llm-scaler","name":"llm-scaler (Intel Arc)","org":"intel","date":"2026-10-16","version":"vllm-1.5","prerelease":false,"releases_30d":1,"recent_tags":["vllm-1.5"],"summary":"","url":"https://github.com/intel/llm-scaler/releases/tag/vllm-1.5","repo_url":"https://github.com/intel/llm-scaler","modality":"","license":""},{"type":"harness","id":"ml-explore/mlx","name":"MLX","org":"ml-explore","date":"2026-10-14","version":"v0.32.4","prerelease":false,"releases_30d":2,"recent_tags":["v0.32.4","v0.32.3"],"summary":"","url":"https://github.com/ml-explore/mlx/releases/tag/v0.32.4","repo_url":"https://github.com/ml-explore/mlx","modality":"","license":""},{"type":"harness","id":"LostRuins/koboldcpp","name":"KoboldCpp","org":"LostRuins","date":"2026-10-13","version":"v1.123.1","prerelease":false,"releases_30d":2,"recent_tags":["v1.123.1","v1.122.1"],"summary":"Download and run the koboldcpp.exe (Windows) or koboldcpp-linux-x64 (Linux), which is a one-file pyinstaller for NVIDIA GPU users.","url":"https://github.com/LostRuins/koboldcpp/releases/tag/v1.123.1","repo_url":"https://github.com/LostRuins/koboldcpp","modality":"","license":""},{"type":"harness","id":"openvinotoolkit/openvino","name":"OpenVINO","org":"openvinotoolkit","date":"2026-10-01","version":"2026.4.1","prerelease":false,"releases_30d":1,"recent_tags":["2026.4.1"],"summary":"You can find the OpenVINO\u2122 toolkit 2026.4.1 release here:","url":"https://github.com/openvinotoolkit/openvino/releases/tag/2026.4.1","repo_url":"https://github.com/openvinotoolkit/openvino","modality":"","license":""}],"ollama_newest":[{"name":"d3","desc":"Turn text, structured data, and images into scored choices, yes/no answers, and ratings.","caps":["vision","decision"],"sizes":["2b","4b","9b","27b"],"updated":"2 days ago","url":"https://ollama.com/library/d3"},{"name":"phonon","desc":"Phonon is Gradium\u2019s text-to-speech family. It turns written text into spoken audio as it generates, making it useful for reading assistant replies, narrating passages, and adding a voice to local apps.","caps":[],"sizes":[],"updated":"6 hours ago","url":"https://ollama.com/library/phonon"},{"name":"minimax-m3.1-flash","desc":"As a next-generation Flash model, M3.1 Flash delivers a generational leap in agentic tasks, coding, Computer Use and cybersecurity, while making online inference nearly three times faster and reducing usage costs by 70%.","caps":["vision","tools","thinking"],"sizes":[],"updated":"5 days ago","url":"https://ollama.com/library/minimax-m3.1-flash"},{"name":"flint","desc":"Flint is a 39.8 billion-parameter mixture-of-experts model from Recursive, built for coding agents.","caps":["tools","thinking"],"sizes":["40b"],"updated":"2 days ago","url":"https://ollama.com/library/flint"},{"name":"d1-omni","desc":"d1-omni is a multimodal 600M decision model from Liquid AI","caps":["vision"],"sizes":["600m"],"updated":"1 week ago","url":"https://ollama.com/library/d1-omni"},{"name":"bonsai-2","desc":"Bonsai 2 from Prism ML is a family of vision models built for ternary weights \u2013 smaller models with close to full-precision quality.","caps":["vision","tools","thinking"],"sizes":["2b","4b","9b","27b"],"updated":"1 week ago","url":"https://ollama.com/library/bonsai-2"},{"name":"beam","desc":"Beam is an open model by Reflection designed to deliver strong performance for areas such as agentic tool use, coding and general reasoning with an emphasis on token efficient usage.","caps":["tools","thinking"],"sizes":[],"updated":"1 week ago","url":"https://ollama.com/library/beam"},{"name":"muse-spark-open","desc":"Muse Spark Open is a natively multimodal model with 816B total parameters (42B active) and a 1M-token context window. Built for coding and agentic work, it excels at long-horizon tasks, complex software engineering, and multi-step reasoning.","caps":["vision","tools","thinking"],"sizes":[],"updated":"1 week ago","url":"https://ollama.com/library/muse-spark-open"},{"name":"pplx-decider","desc":"pplx-decider is a 27B decision model from Perplexity, fine-tuned from Qwen3.8-27B","caps":["vision","decision"],"sizes":["27b"],"updated":"1 week ago","url":"https://ollama.com/library/pplx-decider"},{"name":"d1","desc":"d1 is a 3B decision model from Liquid AI, built on LFM2.5-VL-3B","caps":["vision"],"sizes":["3b"],"updated":"1 week ago","url":"https://ollama.com/library/d1"},{"name":"kev","desc":"Kev scores answers to structured questions about a supplied text state in a single forward pass.","caps":["decision"],"sizes":[],"updated":"1 week ago","url":"https://ollama.com/library/kev"},{"name":"nimble-v3","desc":"A 9B decision model from Bespoke Labs for fast classification, routing, and scoring.","caps":["tools","decision","thinking"],"sizes":["9b"],"updated":"1 week ago","url":"https://ollama.com/library/nimble-v3"},{"name":"nemotron-3.5-bolt","desc":"Nemotron 3.5 Bolt is NVIDIA\u2019s compact vision-language model for local coding, reasoning, and tool use.","caps":["vision","tools","thinking","audio"],"sizes":["e4b"],"updated":"yesterday","url":"https://ollama.com/library/nemotron-3.5-bolt"},{"name":"mistral-large-4","desc":"Mistral Large 4 is Mistral's open-weight, general-purpose multimodal model. Its granular Mixture-of-Experts design activates 52B of 1.05T total parameters, with a 1.6B vision encoder handling images.","caps":["vision","tools","thinking"],"sizes":[],"updated":"2 weeks ago","url":"https://ollama.com/library/mistral-large-4"},{"name":"embeddinggemma-2","desc":"EmbeddingGemma 2 is a multimodal embedding model from Google built on the Gemma 4 architecture.","caps":["vision","embedding","audio"],"sizes":["270m","440m","570m","740m"],"updated":"1 week ago","url":"https://ollama.com/library/embeddinggemma-2"},{"name":"strands-decider","desc":"A 2B decision model from Strands Agents for fast classification, routing, and scoring.","caps":["decision"],"sizes":["2b"],"updated":"1 week ago","url":"https://ollama.com/library/strands-decider"},{"name":"clef-flash","desc":"Clef-Flash is a 9B multimodal model that turns a state and a schema of typed questions into decisions.","caps":["vision","decision"],"sizes":["9b"],"updated":"1 week ago","url":"https://ollama.com/library/clef-flash"},{"name":"clef","desc":"Clef is a 27B multimodal model created by Cloudflare that turns a state and a schema of typed questions into decisions.","caps":["vision","decision"],"sizes":["27b"],"updated":"1 week ago","url":"https://ollama.com/library/clef"},{"name":"laya","desc":"Laya is a 421M decision model from Convai Innovations, fine-tuned from ModernBERT-large.","caps":["embedding","decision"],"sizes":[],"updated":"1 week ago","url":"https://ollama.com/library/laya"},{"name":"tev1","desc":"A 4B decision model from Together AI for fast classification.","caps":["tools","decision","thinking"],"sizes":["0.8b","4b"],"updated":"1 week ago","url":"https://ollama.com/library/tev1"},{"name":"nimble","desc":"A 9B decision model from Bespoke Labs for fast, typed classification.","caps":["tools","decision","thinking"],"sizes":["9b"],"updated":"1 week ago","url":"https://ollama.com/library/nimble"},{"name":"nemotron-3.5-super","desc":"Nemotron 3.5 Super is NVIDIA\u2019s vision-language model for coding, reasoning, tool use, and document understanding","caps":["vision","tools","thinking"],"sizes":["120b"],"updated":"1 week ago","url":"https://ollama.com/library/nemotron-3.5-super"},{"name":"deepseek-v4.1-flash","desc":"Designed for greater capability, faster inference, higher throughput, and scaling to larger models.","caps":["vision","tools","thinking"],"sizes":[],"updated":"1 week ago","url":"https://ollama.com/library/deepseek-v4.1-flash"},{"name":"qwen3.8-flash-next","desc":"This experimental preview of the architecture that will underpin Qwen4.","caps":["vision","tools","thinking"],"sizes":[],"updated":"1 month ago","url":"https://ollama.com/library/qwen3.8-flash-next"},{"name":"glm-5.3-flash","desc":"Z.ai's first natively multimodal model, approaching Claude Opus 4.8 on coding and agentic benchmarks with just 18B active parameters.","caps":["vision","tools","thinking"],"sizes":[],"updated":"1 month ago","url":"https://ollama.com/library/glm-5.3-flash"},{"name":"ornith-1.5","desc":"Chirp Chirp! \ud83d\udc26 We are introducing Ornith-1.5, a major step toward building foundation models through end-to-end self-improvement.","caps":["vision","tools","thinking"],"sizes":["9b","35b","122b","397b"],"updated":"4 days ago","url":"https://ollama.com/library/ornith-1.5"},{"name":"glm-5.3","desc":"Z.ai's flagship model and the most capable open-weights model for coding, with major gains on long-horizon agentic tasks.","caps":["tools","thinking"],"sizes":[],"updated":"1 month ago","url":"https://ollama.com/library/glm-5.3"},{"name":"granite4.2","desc":"IBM Granite Models are a family of enterprise-ready, open foundation models that support multilingual capabilities, coding, retrieval-augmented generation (RAG), tool use, thinking and structured JSON output. Released under Apache 2.0 license.","caps":[],"sizes":["3b","8b","30b"],"updated":"1 month ago","url":"https://ollama.com/library/granite4.2"},{"name":"qwen3.8","desc":"Qwen3.8 delivers substantial gains across coding, professional work, research, and long-horizon agentic tasks.","caps":["vision","tools","thinking"],"sizes":["27b"],"updated":"4 weeks ago","url":"https://ollama.com/library/qwen3.8"},{"name":"nemotron-3.5-lightning","desc":"NVIDIA Nemotron 3.5 Lightning is an open 30B mixture-of-experts (MoE) model with 3B active parameters built for always-on agents.","caps":["tools","thinking"],"sizes":["30b"],"updated":"1 week ago","url":"https://ollama.com/library/nemotron-3.5-lightning"},{"name":"muse-glimmer","desc":"Meta's latest open model built for always-on local agents. 30B parameters, licensed under Apache 2.0 and runs on a single GPU \u2014 tuned for tool use, long tasks, and failure recovery.","caps":["vision","tools","thinking"],"sizes":["30b"],"updated":"1 month ago","url":"https://ollama.com/library/muse-glimmer"},{"name":"kimi-k3","desc":"Kimi K3 is an open-weight, native multimodal agentic model and our most capable model to date.","caps":["vision","tools","thinking"],"sizes":[],"updated":"2 months ago","url":"https://ollama.com/library/kimi-k3"},{"name":"laguna-s-2.1","desc":"Our most capable model to date, designed for long-horizon work. 70.2% on Terminal-Bench 2.1 at 118B-A8B.","caps":["tools","thinking"],"sizes":[],"updated":"1 month ago","url":"https://ollama.com/library/laguna-s-2.1"},{"name":"laguna-xs-2.1","desc":"Laguna XS 2.1 is a 33B total parameter Mixture-of-Experts model with 3B activated parameters per token designed for agentic coding and long-horizon work on a local machine.","caps":["tools","thinking"],"sizes":[],"updated":"1 month ago","url":"https://ollama.com/library/laguna-xs-2.1"},{"name":"ornith","desc":"A self-improving family of open-source models for agentic coding","caps":["tools","thinking"],"sizes":["9b","35b"],"updated":"3 months ago","url":"https://ollama.com/library/ornith"},{"name":"north-mini-code-1.0","desc":"North Mini Code is Cohere's first model for developers \u2014 a 30B Mixture-of-Experts model with 3B active parameters, built for agentic software engineering.","caps":["tools","thinking"],"sizes":[],"updated":"4 months ago","url":"https://ollama.com/library/north-mini-code-1.0"},{"name":"glm-5.2","desc":"GLM-5.2 is Z.ai\u2019s flagship model for the era of long-horizon tasks.","caps":["tools","thinking"],"sizes":[],"updated":"4 months ago","url":"https://ollama.com/library/glm-5.2"},{"name":"kimi-k2.7-code","desc":"Kimi K2.7 Code is Moonshot AI's coding-focused agentic model built upon Kimi K2.6, with substantial improvements on real-world long-horizon coding tasks and roughly 30% lower thinking-token usage.","caps":["vision","tools","thinking"],"sizes":[],"updated":"4 months ago","url":"https://ollama.com/library/kimi-k2.7-code"},{"name":"nemotron-3-ultra","desc":"NVIDIA Nemotron 3 Ultra is built for high-throughput reasoning and long-running agent workflows.","caps":["tools","thinking"],"sizes":[],"updated":"4 months ago","url":"https://ollama.com/library/nemotron-3-ultra"},{"name":"lfm2.5","desc":"LFM2.5-8B-A1B, an edge model built for fast, reliable tool calling on consumer hardware.","caps":["tools","thinking"],"sizes":["8b","24b"],"updated":"1 week ago","url":"https://ollama.com/library/lfm2.5"}],"x_posts":[{"id":"2112047008731308453","user":"Forhanvv","date":"2026-10-19","likes":815,"text":"A new \ud835\udfef\ud835\udfed\ud835\uddd5 \ud835\ude02\ud835\uddfb\ud835\uddf0\ud835\uddf2\ud835\uddfb\ud835\ude00\ud835\uddfc\ud835\uddff\ud835\uddf2\ud835\uddf1 \ud835\uddfa\ud835\uddfc\ud835\uddf1\ud835\uddf2\ud835\uddf9 just dropped for local creative writing and dark fiction.\n\n\ud835\uddea\ud835\uddf6\ud835\uddf0\ud835\uddf8\ud835\uddf2\ud835\uddf1-\ud835\udde7\ud835\ude02\ud835\ude01\ud835\ude02-\ud835\udfef\ud835\udfed\ud835\uddd5 (\ud835\ude03\ud835\udfec.\ud835\udfed) is tuned specifically to eliminate moral lectures, safety preaching, and refusal loops in roleplay:\n\n\u2022 \ud835\udded\ud835\uddf2\ud835\uddff\ud835\uddfc \ud835\udde0\ud835\uddfc\ud835\uddff\ud835\uddee\ud835\uddf9 https://t.co/zhoetC0quy","url":"https://x.com/Forhanvv/status/2112047008731308453","hf":[]},{"id":"2112267555339448352","user":"MiaAI_lab","date":"2026-10-19","likes":675,"text":"I have early access to an *interesting* model. Fits one DGX Spark, and it will be released TOMORROW. \ud83d\udc40","url":"https://x.com/MiaAI_lab/status/2112267555339448352","hf":[]},{"id":"2110893784422908225","user":"ArtificialAnlys","date":"2026-10-16","likes":568,"text":"Meta has released Muse Spark Open, an 816B (42B active) open-weights model that scores 39 on the Artificial Analysis Intelligence Index, making it the #1 US open-weights model\n\nMuse Spark Open (xhigh) scores 22 points above Muse Glimmer (high, 17), Meta's 30B dense open-weights h","url":"https://x.com/ArtificialAnlys/status/2110893784422908225","hf":[]},{"id":"2106579758448336968","user":"pengsonal","date":"2026-10-04","likes":473,"text":"FOUND GLM 5.3 UNCENSORED YOU CAN DOWNLOAD FOR FREE \n\n1. GLM 5.3 Flash Uncensored FP8\n   \u2192 320B total / 18B active\n   \u2192 MIT\n   https://t.co/kPLkN0R9HI\n\n2. GLM 5.3 Uncensored FP8\n   \u2192 full 753B model\n   \u2192 MIT\n   https://t.co/fWdijcfYWG\n\n3. GLM 5.3 Flash Abliterated GGUF\n   \u2192","url":"https://x.com/pengsonal/status/2106579758448336968","hf":[]},{"id":"2111098413907599694","user":"rapidmlx","date":"2026-10-16","likes":385,"text":"\ud83d\ude80 Rapid-MLX 0.16.2 is out! Bringing massive AI power locally to your Apple Silicon Mac.\n\nHighlights of this release: \n\u2728 @NVIDIAAI Nemotron 3.5 (Bolt &amp; Super VL up to 120B) support w/ reasoning &amp; tools! \n\ud83c\udf31 @PrismML Ternary Bonsai 2 support for small-memory Macs!\n\ud83e\uddde @googleg","url":"https://x.com/rapidmlx/status/2111098413907599694","hf":[]},{"id":"2111380259295653976","user":"testingcatalog","date":"2026-10-17","likes":368,"text":"BREAKING \ud83d\udd25: DeepSeek released DeepSeek-V4.1-Flash-1017, a new open-weight model with frontier performance. \n\n&gt; It scores 77% on DeepSWE 1.1 benchmark, outperforming GLM-5.3, Kimi K3 and models like GPT-5.6 Sol and Opus 5. \n\n&gt; The model is now available on DeepSeek Chat and ","url":"https://x.com/testingcatalog/status/2111380259295653976","hf":[]},{"id":"2111540267840901174","user":"MiaAI_lab","date":"2026-10-17","likes":317,"text":"Rejoice single DGX Spark owners!\n\nYou can now run Qwen3.8 Flash Next on TensorFold with the new Nvidia FP4 QAD quant \ud83d\udd25\n\n- 1.1M total FP8 KV cache\n- 256k context, 5 concurrent users\n- 46-55 tok/s on prose, single stream\n- 132 tok/s on prose, 5 concurrent steams\n- Prefill","url":"https://x.com/MiaAI_lab/status/2111540267840901174","hf":[]},{"id":"2112366415575720189","user":"ViC305","date":"2026-10-20","likes":261,"text":"\ud83d\udea8 \ud835\udddc \ud835\udddd\ud835\udde8\ud835\udde6\ud835\udde7 \ud835\uddd4\ud835\uddd6\ud835\udddb\ud835\udddc\ud835\uddd8\ud835\udde9\ud835\uddd8\ud835\uddd7 \ud835\udde6\ud835\udde2\ud835\udde0\ud835\uddd8\ud835\udde7\ud835\udddb\ud835\udddc\ud835\udde1\ud835\uddda \ud835\udddc\ud835\udde1\ud835\uddd6\ud835\udde5\ud835\uddd8\ud835\uddd7\ud835\udddc\ud835\uddd5\ud835\udddf\ud835\uddd8 \ud835\udde2\ud835\udde1 \ud835\udde2\ud835\udde1\ud835\uddd8 \ud835\uddd7\ud835\uddda\ud835\uddeb \ud835\udde6\ud835\udde3\ud835\uddd4\ud835\udde5\ud835\uddde! \ud83e\uddea\ud83d\udd25\nWith the support of @AWERXAI, I'm excited to release something VERY special for the local AI community!\n\n\ud83e\udde0 @deepseek_ai \ud835\uddd7\ud835\uddf2\ud835\uddf2\ud835\uddfd\ud835\udde6\ud835\uddf2\ud835\uddf2\ud835\uddf8 \ud835\udde9\ud835\udff0.\ud835\udfed \ud835\uddd9\ud835\uddf9\ud835\uddee\ud835\ude00\ud835\uddf5 \ud835\udfed\ud835\udfec\ud835\udfed\ud835\udff3\n\ud83d\udce6 https://t.co/RiEkc8wIrw","url":"https://x.com/ViC305/status/2112366415575720189","hf":[]},{"id":"2113508915803832831","user":"MiaAI_lab","date":"2026-10-23","likes":252,"text":"MiniMax-M3.1-Flash open weights are now LIVE \ud83d\udd25\ud83d\udd25\n\n428B params, 23B active.\n\nWeights ~252 GB. Link below. https://t.co/9VHfbbC5Yu","url":"https://x.com/MiaAI_lab/status/2113508915803832831","hf":[]},{"id":"2113994315621867582","user":"TeksEdge","date":"2026-10-24","likes":216,"text":"Qwen4-27B isn't released yet... but the inference stack may have leaked.\ud83d\udc40\n\nA newly merged XPU kernel PR explicitly names:\n\n\ud83d\udc49 Qwen4-27B dense\n\nAnd it's already optimizing the unreleased model for these stats\n\u26a1 FP8 \u2192 2 GPUs\n\ud83e\uddee INT4 \u2192 1 or 2 GPUs\n\ud83d\ude80 MTP \u2192 4 draft tokens\n\ud83e\udde0 QSA https://","url":"https://x.com/TeksEdge/status/2113994315621867582","hf":[]},{"id":"2111615210503168164","user":"Aura_lixx","date":"2026-10-18","likes":164,"text":"50 USEFUL GITHUB REPOS\n\n1. iFixAi \u2014 AI misalignment testing\n\u2192 https://t.co/ZUkRkOOOVi\n\n2. public-apis \u2014 free APIs for everything\n\u2192 https://t.co/Afp9ZQxEIU\u2026\n\n3. build-your-own-x \u2014 learn by building\n\u2192 https://t.co/3In9rFB7jB\u2026\n\n4. developer-roadmap \u2014 learn any tech skill\n\u2192 https://t","url":"https://x.com/Aura_lixx/status/2111615210503168164","hf":[]},{"id":"2112921548910641329","user":"NickDevFE","date":"2026-10-21","likes":149,"text":"Just tested the new MiniMax M3.1 Flash with img2interior! \ud83d\udd25\n\nOne bedroom reference image \u2192 a fully editable Three.js scene.\n\nThe result genuinely impressed me. Beautiful details, lighting, and an incredibly clean interior.\n\nReally excited to see how far we can push this model htt","url":"https://x.com/NickDevFE/status/2112921548910641329","hf":[]},{"id":"2111780423290208277","user":"dealignai","date":"2026-10-18","likes":147,"text":"DeepSeek v4.1 Flash 1017 Fully Uncensored - for you 2x dgx spark people!\n\nI got lazy and cba to make my own exl3 quant atm (wanted to get this out for u guys asap) so go thank @Libertai_DAI for the exl3 2.9bpw quant. In the next few days I\u2019ll be making my own ~2.8-2.9bpw quant ht","url":"https://x.com/dealignai/status/2111780423290208277","hf":[]},{"id":"2111407881585136042","user":"Mekarly","date":"2026-10-17","likes":133,"text":"Been looking into @MetaAI\u2019s new Muse Spark Open, and there are some interesting details here\n\nThe model has 816B total parameters, 42B active parameters and a 1M token context window.\n\nWhat stands out to me is its score of 39 on the @ArtificialAnlys Intelligence Index making it h","url":"https://x.com/Mekarly/status/2111407881585136042","hf":[]},{"id":"2113850470590767267","user":"suraj_sharma14","date":"2026-10-24","likes":131,"text":"Your AI feed is probably noise. Fix it:\n\n1.) @levelsio\n\u27a3 Shipping products fast\n\n2.) @simonw\n\u27a3 Building with LLMs, tools, evals\n\n3.) @DrJimFan\n\u27a3 Agents, robotics, embodied AI\n\n4.) @OpenAI\n\u27a3 Official launches\n\n5.) @HamelHusain\n\u27a3 Evals &amp; LLM ops\n\n6.) @karpathy\n\u27a3 LLM","url":"https://x.com/suraj_sharma14/status/2113850470590767267","hf":[]},{"id":"2113781639146840521","user":"IndependentEco","date":"2026-10-23","likes":129,"text":"Open weight models are in serious trouble.\n\nWhen Kimi K3 dropped months ago it was near frontier in animation and frontend.\n\nAt the time everyone assumed open weight models were closing the gap. \n\nNow K3.1 pro drops and the gap to frontier models like Opus 5.5 or Astra 6 have","url":"https://x.com/IndependentEco/status/2113781639146840521","hf":[]},{"id":"2112389514404462613","user":"ArtificialAnlys","date":"2026-10-20","likes":128,"text":"Ant Group's Ming-3.0-tts is the new #1 open weights model in the Artificial Analysis Controlled Voice TTS Arena and takes the #1 spot on our Pronunciation Robustness leaderboard, level with Fish Audio's Drama 3 Preview and ahead of Eleven v4\n\nMing-3.0-tts is the new open weights ","url":"https://x.com/ArtificialAnlys/status/2112389514404462613","hf":[]},{"id":"2112989805919232060","user":"NVIDIARTXSpark","date":"2026-10-21","likes":112,"text":"DGX Spark is getting new systems, bigger clusters, and easier model setup.\n\nDGX Spark 64GB systems start rolling out Oct. 23, with eight-system cluster support and a new Model Launcher coming later this month. https://t.co/89AsQTw1gq","url":"https://x.com/NVIDIARTXSpark/status/2112989805919232060","hf":[]},{"id":"2113525553114353809","user":"MiaAI_lab","date":"2026-10-23","likes":103,"text":"MiniMax-M3.1-Flash TensorFold recipe for 2x DGX Sparks dropping soon!! EXL3-EAQ quant is ready.\n\nhttps://t.co/3TAzNrMrNQ https://t.co/xOddY3jtmD","url":"https://x.com/MiaAI_lab/status/2113525553114353809","hf":[]},{"id":"2111777166702870615","user":"ivanfioravanti","date":"2026-10-18","likes":101,"text":"MOSS-image + 8 steps LoRA. \nI took inspiration from Jean-Claude post and tried this on DGX Spark:\n- 1024x1024 takes 4.5 secs\n- 2048x2048 takes 17 secs\n\nWHAT A MODEL! @MosiAI_Official \ud83d\ude4f\n\nSome prompts in thread \ud83e\uddf5 https://t.co/iFH52PaJU2 https://t.co/SxMJvRF8me","url":"https://x.com/ivanfioravanti/status/2111777166702870615","hf":[]},{"id":"2113452850122776687","user":"tenstorrent","date":"2026-10-23","likes":97,"text":"Today @MiniMax_AI released the open weights for MiniMax M3.1 Flash \u2013 available now on Tenstorrent Cloud.\n\nExplore in the playground or connect to your agent harness. Start building \u2192 https://t.co/9IL1bl6E6S https://t.co/eZCY02J4LJ","url":"https://x.com/tenstorrent/status/2113452850122776687","hf":[]},{"id":"2112458715223732309","user":"testingcatalog","date":"2026-10-20","likes":83,"text":"DAILY AI BRIEF \ud83d\uddde \u2014 Oct 20\n\nOPENAI \ud83d\udd25:\n- ChatGPT Slides now supports real-time collaborative editing, comments and on-slide edits for Pro, Business and Enterprise plans.\n- OpenAI Platform recently added a \"Decisions\" API playground for testing its decision model with Predicate, htt","url":"https://x.com/testingcatalog/status/2112458715223732309","hf":[]},{"id":"2107209226481479964","user":"mercor","date":"2026-10-05","likes":83,"text":"Today Reflection released Beam, their first open-weight model.\n\nBeam was trained with a particular focus on coding and agentic performance, advancing the Western open-weights frontier.\n\nCongratulations to the @reflection_ai team. https://t.co/7a1FRjEKJl","url":"https://x.com/mercor/status/2107209226481479964","hf":[]},{"id":"2113230869699637548","user":"MAXdeg0","date":"2026-10-22","likes":82,"text":"MINIMAX M3.1 FLASH vs ASTRA SOL 6.1 \n\nTested with the exact same prompt.\n\nThe task: build a cinematic 3D Earth scene with a space shuttle in low orbit, heat shield tiles, volumetric clouds, city lights on the dark side, and sunlight reflecting off the metal.\n\n- MiniMax M3.1 https","url":"https://x.com/MAXdeg0/status/2113230869699637548","hf":[]},{"id":"2113985781928800510","user":"vikktorrrre","date":"2026-10-24","likes":81,"text":"Anthropic might have to release Fable 5.5 next week\n\nOver the past 12 hours, Two Core members who contributed to the growth of Claude Code have left Anthropic, and as you would expect, this doesn't bode well, considering they plan on going public in a few weeks with their IPO. ht","url":"https://x.com/vikktorrrre/status/2113985781928800510","hf":[]},{"id":"2111344167997558857","user":"Presidentlin","date":"2026-10-17","likes":80,"text":"I really missed the hours before a DeepSeek release.\n\nIt's one of the last remaining joys I have in this space.\n\nOther Open Weights models first tease, then ship a week or month later.\n\nUS Closed Source ship the best state of the model, collect the PR and slightly nerf the model","url":"https://x.com/Presidentlin/status/2111344167997558857","hf":[]},{"id":"2112386229564285133","user":"NeoAIForecast","date":"2026-10-20","likes":78,"text":"One DGX Spark. DeepSeek V4.1 Flash 1017. 1.62 bits per expert.\n\n@ViC305 just dropped a SAGE-EXL3 quant and a one-Spark vLLM recipe!!!\n\n&gt;35.5 tok/s decode\n&gt;1,189 tok/s max prefill\n&gt;229K context so far\n&gt;99.61% answer-token agreement on HumanEval vs the native checkpoint","url":"https://x.com/NeoAIForecast/status/2112386229564285133","hf":[]},{"id":"2112822741208314258","user":"MiaAI_lab","date":"2026-10-21","likes":78,"text":"Run Ornith-1.5-122B on your DGX Spark \u2728\n\nUses my Ornith-1.5-122B-EXL3-EAQ 4 bpw quant. Recipe can easily be adapted for the RTX PRO 6000 too.\n\n- Runs on TensorFold\n- 256K context, total 1M KV cache\n- 4 concurrent streams by default\n- 86 GB total memory\n- Uses DFlash2 drafts","url":"https://x.com/MiaAI_lab/status/2112822741208314258","hf":[]},{"id":"2113560791941091619","user":"testingcatalog","date":"2026-10-23","likes":75,"text":"MiniMax has published MiniMax M3.1 Flash weights on HuggingFace. \n\n&gt; The model has 428B parameters total, 23B activated per token.\n\n&gt; MiniMax M3.1 Flash, also known as \"Space Bunny\" stealth model from OpenRouter, was announced on October 19 and is now available on MiniMax h","url":"https://x.com/testingcatalog/status/2113560791941091619","hf":[]},{"id":"2111815926534516928","user":"KBlueleaf","date":"2026-10-18","likes":73,"text":"KohakuMossTrainer is released:\nhttps://t.co/Ga4Hpjp4C1\n\n**Features**\n- LyCORIS all algorithms\n- SM70 ~ SM120\n- Intel Arc GPU\n- Windows and LInux\n- multi GPU training\n- Model offload\n\n***Try to test it now!***","url":"https://x.com/KBlueleaf/status/2111815926534516928","hf":[]},{"id":"2113402152420827494","user":"MiaAI_lab","date":"2026-10-22","likes":67,"text":"MiniMax-M3.1-Flash open weights drop tomorrow. https://t.co/f2Khbr8Kou","url":"https://x.com/MiaAI_lab/status/2113402152420827494","hf":[]},{"id":"2112113265056567398","user":"SPAC89","date":"2026-10-19","likes":64,"text":"Space Bunny is officially MiniMax 3.1, and it's expected to be even better than DeepSeek V4.1 Flash 1017! When it was available for free, it became the most used model on OpenCode and OpenRouter, so there's a good chance we could see both MiniMax 3.1 Flash and MiniMax 3.1","url":"https://x.com/SPAC89/status/2112113265056567398","hf":[]},{"id":"2111523015234384166","user":"ViC305","date":"2026-10-17","likes":63,"text":"\ud83d\udea8 @nvidiaai FP4-LOCAL: THE GOOD, THE BAD &amp; THE UGLY! \ud83d\udd25\n\nI tested Qwen3.8-Flash-Next 4 ways. FP4-Local TP1 &amp; 2, NVFP4 and EXL3\n\nFP4: 1 Spark: 90.8% SixCat, 68 tok/s. 2 Sparks: 32% slower decode. DIY quantization? That's the ugly part.\n\nMy full article \ud83d\udc47 https://t.co/5GbiEd","url":"https://x.com/ViC305/status/2111523015234384166","hf":[]},{"id":"2113948624359166103","user":"niklaslenz_ai","date":"2026-10-24","likes":60,"text":"DeepSeek-V4.1-Flash-1017 on a Mac Studio M5 Ultra 256 GB: 54.5 tok/s on antirez's ds4.\n\nantirez's Q2 GGUF of the model, in kernelpool's fork of his ds4 engine (commit 72ccf36), in ds4's batched mode with 5 sessions. The 152 GiB of main weights stay in memory, the 189 GiB of https","url":"https://x.com/niklaslenz_ai/status/2113948624359166103","hf":[]},{"id":"2110883444880216427","user":"thepirateface","date":"2026-10-16","likes":59,"text":"\ud83e\uddf2 Community torrents \u00b7 16 Oct 2026\n\n1000 model torrents added:\n\u2022 autotrust/GEV-26B-Decide-NVFP4\n\u2022 bartowski/FrogNano-4B-2609-GGUF\n\u2022 mradermacher/gemma-4-31B-it-The-DECKARD-HERETIC-UNCENSORED-Thinking-i1-GGUF\n\u2022 huihui-ai/Huihui-GLM-5.3-Flash-abliterated-GGUF\n\u2022","url":"https://x.com/thepirateface/status/2110883444880216427","hf":[]},{"id":"2113111853740294213","user":"superalesha","date":"2026-10-22","likes":57,"text":"A new US lab just shipped a real open model.\n\nRecursive Flint. 40B coding MoE, 2.5B active, trained from scratch in under 4 months. Apache 2.0, weights on Hugging Face from day 1.\n\n125+ tok/s on a MacBook Pro. 150+ on RTX Spark. About 7 GiB of KV cache at 262K context.\n\nThey http","url":"https://x.com/superalesha/status/2113111853740294213","hf":[]},{"id":"2112960579979579631","user":"ivanfioravanti","date":"2026-10-21","likes":56,"text":"NVIDIA released Flare, a local inference engine for DGX Spark. I'll surely give it a try!!! \n\nBut I've seen some trade-off on quality or strange default settings so far:\n- FP8 KV cache by default, even for checkpoints that don't ask for it \ud83d\udc40\n- No TP=3: every recipe supports only ","url":"https://x.com/ivanfioravanti/status/2112960579979579631","hf":[]},{"id":"2113283760783544415","user":"hasantoxr","date":"2026-10-22","likes":56,"text":"For 10 days the #1 model on OpenRouter was a mystery called Space Bunny.\n\n56.7 trillion tokens. Nobody knew who made it. Most guesses said Kimi.\n\nMonday the mask came off. It was MiniMax M3.1 Flash.\n\n$0.15 per million input tokens.\n\nSo I gave it a real job \ud83e\uddf5","url":"https://x.com/hasantoxr/status/2113283760783544415","hf":[]},{"id":"2112955241780154730","user":"CaimingXiong","date":"2026-10-21","likes":51,"text":"At Recursive, we\u2019re exploring the next generation of AI training.\n\nToday, we\u2019re releasing our first model: Recursive Flint, a state-of-the-art local model for coding agents, built with our automated research system. Its open weights are now available on Hugging Face.\n\nFlint marks","url":"https://x.com/CaimingXiong/status/2112955241780154730","hf":[]},{"id":"2110782204670627846","user":"notjazii","date":"2026-10-15","likes":49,"text":"it\u2019s not looking good for chinese ai models\n\nmeta just released open weights for muse spark and it beats almost every other open weight model\n\nmicrosoft is also about to finish pre training its new beast that beats all chinese models\n\nand all the other us models are way ahead htt","url":"https://x.com/notjazii/status/2110782204670627846","hf":[]}],"devices":[{"key":"nv8","group":"NVIDIA","label":"8GB GPU (4060 / 3070)","usable_gb":7,"arc":false,"discrete":true,"slow":false,"lab":false,"mem_gb":8},{"key":"gpu12","group":"NVIDIA","label":"12GB GPU (RTX 3060)","usable_gb":11,"arc":false,"discrete":true,"slow":false,"lab":true,"mem_gb":12},{"key":"gpu16","group":"NVIDIA","label":"16GB GPU (5060 Ti / 4060 Ti / 4080 / 5080)","usable_gb":15,"arc":false,"discrete":true,"slow":false,"lab":true,"mem_gb":16},{"key":"nv24","group":"NVIDIA","label":"24GB GPU (3090 / 4090)","usable_gb":23,"arc":false,"discrete":true,"slow":false,"lab":false,"mem_gb":24},{"key":"nv32","group":"NVIDIA","label":"32GB GPU (5090)","usable_gb":31,"arc":false,"discrete":true,"slow":false,"lab":false,"mem_gb":32},{"key":"nv48","group":"NVIDIA","label":"48GB GPU (RTX 6000 Ada / A6000)","usable_gb":47,"arc":false,"discrete":true,"slow":false,"lab":false,"mem_gb":48},{"key":"nv96","group":"NVIDIA","label":"96GB GPU (RTX Pro 6000 Blackwell)","usable_gb":94,"arc":false,"discrete":true,"slow":false,"lab":false,"mem_gb":96},{"key":"spark1","group":"NVIDIA","label":"1\u00d7 DGX Spark (128GB)","usable_gb":115,"arc":false,"discrete":false,"slow":false,"lab":true,"mem_gb":128},{"key":"spark2","group":"NVIDIA","label":"2\u00d7 DGX Spark (256GB)","usable_gb":230,"arc":false,"discrete":false,"slow":false,"lab":true,"mem_gb":256},{"key":"amd16","group":"AMD","label":"16GB GPU (7800 XT / 9070 XT)","usable_gb":15,"arc":false,"discrete":true,"slow":false,"lab":false,"mem_gb":16},{"key":"amd24","group":"AMD","label":"24GB GPU (7900 XTX)","usable_gb":23,"arc":false,"discrete":true,"slow":false,"lab":false,"mem_gb":24},{"key":"strix","group":"AMD","label":"Strix Halo 128GB unified (~96GB usable)","usable_gb":96,"arc":false,"discrete":false,"slow":false,"lab":false,"mem_gb":128},{"key":"gpu24","group":"Intel Arc","label":"Arc B60 24GB","usable_gb":23,"arc":true,"discrete":true,"slow":false,"lab":true,"mem_gb":24},{"key":"gpu32","group":"Intel Arc","label":"Arc B70 32GB","usable_gb":31,"arc":true,"discrete":true,"slow":false,"lab":true,"mem_gb":32},{"key":"mac16","group":"Apple","label":"Mac 16GB","usable_gb":12,"arc":false,"discrete":false,"slow":false,"lab":false,"mem_gb":16},{"key":"mac24","group":"Apple","label":"Mac 24GB","usable_gb":18,"arc":false,"discrete":false,"slow":false,"lab":false,"mem_gb":24},{"key":"mac36","group":"Apple","label":"Mac 36GB","usable_gb":27,"arc":false,"discrete":false,"slow":false,"lab":false,"mem_gb":36},{"key":"mac48","group":"Apple","label":"Mac 48GB","usable_gb":36,"arc":false,"discrete":false,"slow":false,"lab":false,"mem_gb":48},{"key":"mac64","group":"Apple","label":"Mac 64GB (M1 Max)","usable_gb":48,"arc":false,"discrete":false,"slow":false,"lab":true,"mem_gb":64},{"key":"mac96","group":"Apple","label":"Mac 96GB","usable_gb":72,"arc":false,"discrete":false,"slow":false,"lab":false,"mem_gb":96},{"key":"mac128","group":"Apple","label":"Mac 128GB","usable_gb":96,"arc":false,"discrete":false,"slow":false,"lab":false,"mem_gb":128},{"key":"mac192","group":"Apple","label":"Mac 192GB","usable_gb":144,"arc":false,"discrete":false,"slow":false,"lab":false,"mem_gb":192},{"key":"mac256","group":"Apple","label":"Mac 256GB","usable_gb":192,"arc":false,"discrete":false,"slow":false,"lab":false,"mem_gb":256},{"key":"mac512","group":"Apple","label":"Mac 512GB","usable_gb":384,"arc":false,"discrete":false,"slow":false,"lab":false,"mem_gb":512},{"key":"cpu32","group":"CPU only","label":"CPU only, 32GB RAM (slow)","usable_gb":26,"arc":false,"discrete":false,"slow":true,"lab":false,"mem_gb":32},{"key":"cpu64","group":"CPU only","label":"CPU only, 64GB RAM (slow)","usable_gb":56,"arc":false,"discrete":false,"slow":true,"lab":false,"mem_gb":64},{"key":"cpu128","group":"CPU only","label":"CPU only, 128GB RAM (slow)","usable_gb":118,"arc":false,"discrete":false,"slow":true,"lab":false,"mem_gb":128},{"key":"cpu256","group":"CPU only","label":"CPU only, 256GB RAM (slow)","usable_gb":240,"arc":false,"discrete":false,"slow":true,"lab":false,"mem_gb":256}],"estimate_note":"Need = weight size + FP16 KV cache for 8192 tokens (from config.json when available, otherwise a rough 8% of Q8 weights) + 1.0GB overhead. Weight size comes from the real quant files on Hugging Face when they exist (Q4_K_M, Q8_0 and BF16/F16 GGUF from the official org, unsloth, bartowski, ggml-org, lmstudio-community or mradermacher; MLX 4-bit/8-bit/bf16; or the official BF16 safetensors; another community GGUF only as a last resort, marked community), checked against the parameter count. When no matching file exists it is ESTIMATED at ~4.8/8.5/16 bits per weight and labelled 'est'. Q4 may use Unsloth UD-Q4_K_XL or Q4_K_S when no Q4_K_M exists. Usable memory per device is assumed below physical (GPUs \u2248 VRAM \u2212 1GB; Mac \u224875% of unified RAM; Strix Halo \u224896GB; CPU = RAM minus OS, and slow). est.fit in the JSON is for one card per device; the page can multiply discrete GPUs by 2 or 4, which assumes the runtime splits the model across cards (tensor/pipeline parallel or llama.cpp layer split). Arc classes only count Q8/FP16, matching what performs well on Arc. MoE models still need all weights resident; active params only affect speed. Fit is still a sizing calculation, not verified by running."}