Agnes-AI/Agnes-3.0-Qwen
archived · low-signalmodel
agnes-3.0-qwen is a post-trained derivative of Qwen/Qwen3.8-Flash-Next. Mixture-of-experts, 180B total (512 experts, 10 per token). Context: 1M tokens (config). License: qwen-community-1.0. Vendor-reported: "Reliable instruction following: on IFBench, agnes-3.0-qwen outperforms GPT-5.6 Sol (Max) and Claude Opus 4.8 (Max)."
- Vendor-reported: Reliable instruction following: on IFBench, agnes-3.0-qwen outperforms GPT-5.6 Sol (Max) and Claude Opus 4.8 (Max).
- Org
- Agnes-AI
- Released
- 2026-10-10
- Status
- archived · low-signal
- License
- qwen-community-1.0 (restricted)
- Parameters
- 180.0B
- Architecture
- MoE · 512 experts, 10/token · Qwen4ExpForConditionalGeneration
- Context
- 1,048,576 tokens
- Task
- image-text-to-text · vision
- Signal
- 18 HF likes · 269 downloads
- Archived
- 2026-10-24 (no longer listed by the collector; stats as last seen)
Runs on
real quant files; need = weights + FP16 KV cache for 8K tokens + 1GB. A sizing calculation, not tested.
- Q4 ~128.9 GB: Mac 192GB, Mac 256GB
- Q8 ~192.8 GB: 2× DGX Spark (256GB), CPU only, 256GB RAM (slow)
- FP16 ~361.0 GB: Mac 512GB
Links
First published here 2026-10-10 · updated 2026-10-24 · markdown · JSON