Mistral Small 4.
Mistral AI Mar 2026 Open weights Small
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AA-LCR (Artificial Analysis Long Context Reasoning) | 71.2% | #6 of 31 | Model card | |
| MMLU-Pro | 78% | #87 of 151 | Model card | |
| COLLIE | 62.9% | #8 of 12 | Model card | |
| Arena-Hard-Auto | 58.3% | #30 of 39 | Model card | |
| Longform Creative Writing (EQ-Bench) | 43.2 | #94 of 119 | Official leaderboard | |
| SEA-HELM | 52.1 | #54 of 62 | Official leaderboard | |
| IFBench | 48% | #61 of 70 | Model card | |
| Judgemark v2.1 | 51.6 | #38 of 42 | Official leaderboard | |
| BullshitBench V2 | 6.00% | #116 of 122 | Official leaderboard | |
| Judgemark v4 | 16.5 | #39 of 40 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AIME 2025 | 83.8% | #77 of 143 | Model card | |
| GPQA Diamond | 71.2% | #166 of 284 | Model card | |
| DTBench | 51.5% | #110 of 178 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench | 63.6% | #32 of 73 | Model card | |
| StructEval | 72.5% | #17 of 28 | Official leaderboard | |
| ALE-Bench | 498 | #100 of 121 | Epoch AI | |
| Vibe Code Bench | 0% | #114 of 115 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| PinchBench | 71.3% | #20 of 48 | Official leaderboard | |
| PinchBench 2.0 | 64.6% | #39 of 57 | Official leaderboard | |
| GAIA | 17.6% | #48 of 57 | Official leaderboard | |
| Pencil Puzzle Bench | 0% | #48 of 53 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMMU-Pro | 60% | #56 of 72 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Human Pathogen Capabilities Test v2 (HPCT v2) | 27.5% | #67 of 71 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LMCA | 20.6% | #104 of 139 | Epoch AI | |
| Conceptual Reasoning Index | 30.5 | #113 of 149 | Official leaderboard | |
| ACCoRD | 39.0 | #132 of 150 | Official leaderboard | |
| Political Science LLM Benchmark | 62.9% | #26 of 29 | Official leaderboard | |
| SAGE v1.1 (Vals) | 59.2% | #28 of 28 | Official leaderboard |
The benchmarks the lab chose to headline, with the scores it gave.