JIRU-W/e32_top8_seq2048_gbs128_10b_perlayeraux
just landedmodel
Text generation model from JIRU-W (HF pipeline tag; the card has no one-line description). Mixture-of-experts, 3.7B total (32 experts, 8 per token). Context: 4K tokens (config). License: apache-2.0.
- Org
- JIRU-W
- Released
- 2026-10-29
- Status
- just landed
- License
- apache-2.0 (permissive)
- Parameters
- 3.7B
- Architecture
- MoE · 32 experts, 8/token · MaskMoeForCausalLM
- Context
- 4,096 tokens
- Task
- text-generation · text
- Signal
- 1 HF likes · 5165 downloads
Runs on
mixed: real files + estimate; need = weights + FP16 KV cache for 8K tokens + 1GB. A sizing calculation, not tested.
- Q4 ~4.3 GB (estimated size): no listed device
- Q8 ~6.0 GB (estimated size): 8GB GPU (4060 / 3070)
- FP16 ~9.5 GB: 12GB GPU (RTX 3060), 16GB GPU (5060 Ti / 4060 Ti / 4080 / 5080), 24GB GPU (3090 / 4090), 32GB GPU (5090), 48GB GPU (RTX 6000 Ada / A6000), 96GB GPU (RTX Pro 6000 Blackwell), 1× DGX Spark (128GB), 2× DGX Spark (256GB), 16GB GPU (7800 XT / 9070 XT), 24GB GPU (7900 XTX), Strix Halo 128GB unified (~96GB usable), Arc B60 24GB, Arc B70 32GB, Mac 16GB, Mac 24GB, Mac 36GB, Mac 48GB, Mac 64GB (M1 Max), Mac 96GB, Mac 128GB, Mac 192GB, Mac 256GB, Mac 512GB, CPU only, 32GB RAM (slow), CPU only, 64GB RAM (slow), CPU only, 128GB RAM (slow), CPU only, 256GB RAM (slow)
Links
First published here 2026-10-29 · updated 2026-10-29 · markdown · JSON