Loading…
Loading…
Models / DeepSeek-V4-Pro
1 model (~1.6e+03B) · ↓ 382.9K downloads a month · from $44.03/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 |
|---|---|---|---|---|
| Originaldeepseek-ai/DeepSeek-V4-Pro | vLLM | 865 GB | 964.2 GB | $44.03/h 8x Nvidia H200 |
| NVFP4 (nvidia)nvidia/DeepSeek-V4-Pro-NVFP4 | vLLM | 913 GB | 1017.9 GB | $44.03/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").