Loading…
Loading…
Models / Hy4-preview
2 models (~780B to ~804B) · ↓ 33.9K downloads a month · from $52.18/h on 8x Nvidia H200 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 |
|---|---|---|---|---|
| Originaltencent/Hy4-preview | vLLM | 1560 GB | 1737 GB | None big enough now |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| Originaltencent/Hy4-preview-FP8 | vLLM | 814 GB | 907.9 GB | $52.18/h 8x Nvidia H200 |
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").