Loading…
Loading…
Models / Mistral
GPU memory needed (smallest version of each): Mistral 22B about 53.7 GB. The cheapest runs on 4x RTX 3090 from $0.94/h right now.
2 models (22B to 22B) · ↓ 87 downloads a month · from $0.94/h on 4x RTX 3090 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 |
|---|---|---|---|---|
| OriginalVezora/Mistral-22B-v0.2 | vLLM | 44 GB | 53.7 GB | $0.94/h 4x RTX 3090 |
| Version | Runtime | Weights | GPU memory | Cheapest machine now |
|---|---|---|---|---|
| OriginalVezora/Mistral-22B-v0.1 | vLLM | 44 GB | 53.7 GB | $0.94/h 4x RTX 3090 |
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").