Loading…
Loading…
Models / Qwen3.8-2.4T-A95B
1 model (~2.45e+03B) · ↓ 64.9K downloads a month. 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 |
|---|---|---|---|---|
| OriginalQwen/Qwen3.8-2.4T-A95B | vLLM | 4892 GB | 5441.6 GB | None big enough now |
| FP8Qwen/Qwen3.8-2.4T-A95B-FP8 | vLLM | 2496 GB | 2779.1 GB | None big enough now |
| Quark MXFP4 (amd)amd/Qwen3.8-2.4T-A95B-Quark-MXFP4 | vLLM | 1372 GB | 1530.5 GB | None big enough now |
| NVFP4 (nvidia)nvidia/Qwen3.8-2.4T-A95B-NVFP4 | vLLM | 1444 GB | 1610.6 GB | None big enough now |
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").