AI model maker
Mistral AI
72 models on Artificials, 24 with published test results and 56 with open weights you can download.
Where Mistral AI stands
Its best model’s rank among all models tested on each skill area’s main test, the same tests model pages use. A longer bar means a better rank.
- OverallMistral Medium 3.5 on Epoch Capabilities IndexTop 41%
- CodingMistral Medium 3.5 on Arena codingTop 26%
- MathMistral Medium 3 on Mock AIME 2024–2025#129 of 198
- LanguageMistral Medium 3.5 on Arena creative writingTop 25%
- ReasoningMistral Medium 3 on GPQA Diamond#126 of 222
On the capability index
Its best 12 of 16 models on the Epoch Capabilities Index, each with its rank among all 274 models. The top score is 167.3, held by Claude Opus 5.5 from Anthropic.
Every test
Its best result on each published test, and that result’s rank among every model tested. Scores from different tests are never combined.
| Test | Best model | Score | Rank |
|---|---|---|---|
| Epoch Capabilities Index | Mistral Medium 3.5 | 141.3 | #110 of 274 |
| Arena instruction following | Mistral Medium 3.5 | 1422 | #92 of 389 |
| Arena creative writing | Mistral Medium 3.5 | 1400 | #95 of 387 |
| Arena math | Mistral Medium 3.5 | 1429 | #93 of 372 |
| Arena coding | Mistral Medium 3.5 | 1479 | #97 of 384 |
| Arena text | Mistral Medium 3.5 | 1427 | #99 of 389 |
| Arena hard prompts | Mistral Medium 3.5 | 1446 | #105 of 389 |
| MATH Level 5 | Mistral Medium 3 | 81.6% | #29 of 97 |
| GPQA Diamond | Mistral Medium 3 | 59.5% | #126 of 222 |
| Mock AIME 2024–2025 | Mistral Medium 3 | 32.2% | #129 of 198 |
| SWE-bench Verified, bash only | Devstral 2 | 53.8% | #28 of 42 |
| Chess puzzles | Magistral Small 1.2 | 3.0% | #108 of 141 |
| Arena WebDev | Mistral Medium 3.5 | 1263 | #107 of 123 |
Every model
Newest first. The price is per 1M tokens, blended three input tokens to one output token: Mistral AI’s own listing, or else the middle price across hosts.
| Model | Released | Context | Price | Capability index |
|---|---|---|---|---|
| Mistral Large 4 | Oct 6, 2026 | 524.3K tokens | $1.03 | Not on the index |
| Mistral Large 4 (0) | Oct 6, 2026 | 524.3K tokens | $1.03 | Not on the index |
| Mistral Medium 3.5 | Apr 29, 2026 | 262.1K tokens | $3.00 | 141.3 |
| Mistral Medium | Apr 29, 2026 | 262.1K tokens | $3.00 | Not on the index |
| Mistral Medium 3.5 | Apr 29, 2026 | 262.1K tokens | $3.00 | Not on the index |
| Mistral Medium 3.5 (128b) | Apr 29, 2026 | 262.1K tokens | $3.02 | Not on the index |
| Mistral Small | Mar 16, 2026 | 262.1K tokens | $0.263 | Not on the index |
| Mistral Small 4 | Mar 16, 2026 | 262.1K tokens | $0.263 | Not on the index |
| Mistral Small 4 (119b 2603) | Mar 16, 2026 | 262.1K tokens | $0.42 | Not on the index |
| Voxtral Mini TTS | Mar 1, 2026 | Not reported | No paid listing | Not on the index |
| Voxtral Mini | Feb 1, 2026 | Not reported | No paid listing | Not on the index |
| Devstral 2 | Dec 9, 2025 | 256K tokens | $0.80 | Not on the index |
| Devstral 2 | Dec 9, 2025 | 262.1K tokens | $0.80 | Not on the index |
| Devstral 2 | Dec 9, 2025 | 262.1K tokens | $0.80 | Not on the index |
| Devstral 2 (123b instruct 2512) | Dec 9, 2025 | 262.1K tokens | $0.65 | Not on the index |
Models, prices, context windows and release dates from models.dev (MIT). Test results from Epoch AI (CC BY 4.0), Arena (CC BY 4.0) and SWE-bench (CC BY-NC 4.0). A model counts here when Mistral AI sells it, several hosts list it or a test covers it. Each figure is the source’s own; nothing is combined into a new score.