Mistral cloud models
32 models tracked via Artificial Analysis. Compare cloud performance, then find local GGUF versions in the GraySoft model catalog.
| Model | Intelligence | Speed (tok/s) |
|---|---|---|
| Mistral Medium 3.5 | 30.4 | 124.888 |
| Mistral Small 4 (Reasoning) | 19.7 | 152.784 |
| Devstral 2 | 19.2 | 126.105 |
| Magistral Medium 1.2 | 18 | 0 |
| Devstral Small 2 | 17.7 | 124.802 |
| Mistral Large 3 | 15.9 | 72.607 |
| Mistral Small 3.1 | 14.9 | 0 |
| Mistral Medium 3.1 | 14.7 | 0 |
| Mistral Medium 3 | 12.5 | 0 |
| Magistral Medium 1 | 12.5 | 0 |
| Devstral Medium | 12.4 | 0 |
| Mistral Small 4 (Non-reasoning) | 12.3 | 144.096 |
| Devstral Small (May '25) | 11.8 | 0 |
| Magistral Small 1.2 | 11.5 | 0 |
| Ministral 3 14B | 11.2 | 96.957 |
| Mistral Small 3.2 | 10.7 | 0 |
| Magistral Small 1 | 10.6 | 0 |
| Devstral Small (Jul '25) | 9.1 | 0 |
| Ministral 3 8B | 9 | 113.539 |
| Mistral Large 2 (Nov '24) | 9 | 0 |
| Pixtral Large | 8 | 0 |
| Ministral 3 3B | 7.1 | 208.136 |
| Mistral Large 2 (Jul '24) | 7 | 0 |
| Mistral Small 3 | 6.7 | 0 |
| Mistral Saba | 6.2 | 0 |
| Mistral Small (Sep '24) | 4.3 | 0 |
| Mistral Large (Feb '24) | 4.1 | 0 |
| Mixtral 8x22B Instruct | 4 | 0 |
| Mistral Small (Feb '24) | 3.2 | 0 |
| Mistral Medium | 3.2 | 0 |
| Mixtral 8x7B Instruct | 2 | 0 |
| Mistral 7B Instruct | 1.7 | 0 |
Run models locally with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.