Loading…
Loading…
Models / Huihui-Qwen3.6
2 models (35B to 35B) · ↓ 77.5K downloads a month · from $0.44/h on RTX 5000 Ada right now. Each model's versions do the same job; quantized ones (AWQ, FP8, 4-bit) need less GPU memory and run on cheaper machines.
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| FP8 DYNAMIC (coolthor)coolthor/Huihui-Qwen3.6-35B-A3B-abliterated-FP8-DYNAMIC | vLLM | 38 GB | 45.6 GB | $0.46/h RTX A6000 |
| NVFP4 (YuYu1015)YuYu1015/Huihui-Qwen3.6-35B-A3B-abliterated-NVFP4 | vLLM | 25 GB | 30.8 GB | $0.44/h RTX 5000 Ada |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| Originalhuihui-ai/Huihui-Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated | vLLM | 72 GB | 82.9 GB | $0.89/h 2x Nvidia A40 |
Model details come from Hugging Face and are refreshed daily; prices follow the live offers. GPU memory is GridPup's estimate: the weights, room for an 8,192-token conversation (chat models) and the runtime's own memory. Large models are split across the GPUs of one machine ("4x RTX 4090").