Loading…
Loading…
Models / Mistral-Nemo-Base-2407
1 model (~12.2B) · ↓ 11.7K downloads a month · from $0.13/h on RTX A4000 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 |
|---|---|---|---|---|
| mistral nemo instruct 2407 awq (casperhansen)casperhansen/mistral-nemo-instruct-2407-awq | vLLM | 8.3 GB | 13 GB | $0.13/h RTX A4000 |
| Onnx (Prince-1)Prince-1/Mistral-Nemo-Base-2407-Onnx | ONNX Runtime GenAI | 25 GB | 31.1 GB | $0.25/h Nvidia V100 |
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").