Mistral · model
Mistral Large 4 benchmark scores
As of 2026-10-07, Mistral Large 4 (Mistral) has measured scores on 12 benchmarks and estimated scores on 85 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 90.8% | estimated ± 6.8 pp, medium confidence |
| AA-HLE | 35.0% | measured |
| AA-SciCode | 54.2% | measured |
| Artificial Analysis Intelligence Index | 38.4% | measured |
| CritPt | 10.6% | measured |
| GDPval-AA | 46.2% | measured |
| GPQA Diamond | 89.2% | estimated ± 2.2 pp, high confidence |
| HLE | 44.4% | estimated ± 9.5 pp, medium confidence |
| AA-LCR | 81.3% | measured |
| LiveCodeBench v6 | 95.2% | estimated ± 3.3 pp, medium confidence |
| AA-Omniscience Accuracy | 25.8% | measured |
| SciCode | 53.8% | estimated ± 3.2 pp, high confidence |
| Terminal-Bench 2.1 | 87.7% | estimated ± 3.9 pp, high confidence |
| BrowseComp | 82.7% | estimated ± 10.1 pp, low confidence |
| HLE w/ tools | 55.3% | estimated ± 6.1 pp, medium confidence |
| LongBench v2 | 61.5% | estimated ± 5.9 pp, low confidence |
| AA Coding Index | 70.6% | estimated ± 6.4 pp, medium confidence |
| AA-MMMU-Pro | 76.4% | measured |
| APEX-Agents | 38.6% | estimated ± 10.0 pp, low confidence |
| APEX-Agents-AA | 34.6% | estimated ± 9.1 pp, medium confidence |
| BioMysteryBench (human-difficult) | 31.9% | estimated ± 5.9 pp, low confidence |
| DeepSearchQA | 90.1% | estimated ± 8.4 pp, medium confidence |
| SWE-bench Verified | 86.4% | estimated ± 4.4 pp, medium confidence |
| AutomationBench | 45.5% | estimated ± 7.6 pp, low confidence |
| BFCL v4 | 77.1% | estimated ± 2.6 pp, medium confidence |
| CyberGym | 78.8% | estimated ± 6.2 pp, medium confidence |
| JobBench | 47.3% | estimated ± 8.6 pp, medium confidence |
| SWE-bench Pro | 65.7% | estimated ± 6.8 pp, medium confidence |
| WideResearch | 96.5% | estimated ± 9.1 pp, low confidence |
| AA Agentic Index | 41.4% | estimated ± 2.8 pp, high confidence |
| MMLU-Pro | 85.6% | estimated ± 3.5 pp, high confidence |
| GPQA | 91.8% | estimated ± 3.2 pp, high confidence |
| Gert Labs | 69.3% | estimated ± 8.9 pp, medium confidence |
| AA-AnalystAgent | 44.0% | estimated ± 6.5 pp, medium confidence |
| AA EnterpriseOps-Gym | 43.6% | estimated ± 4.4 pp, high confidence |
| ARC-AGI-1 | 92.6% | estimated ± 5.4 pp, medium confidence |
| ARC-AGI-2 | 71.1% | estimated ± 14.2 pp, low confidence |
| CursorBench 3.1 | 54.4% | estimated ± 3.4 pp, medium confidence |
| CursorBench 3.2 | 60.4% | estimated ± 5.3 pp, medium confidence |
| FrontierCode 1.1 Main | 47.7% | estimated ± 4.3 pp, low confidence |
| FrontierSWE v2 | 24.6% | estimated ± 14.6 pp, low confidence |
| OfficeQA Pro | 50.3% | estimated ± 7.2 pp, medium confidence |
| OSWorld-Verified | 79.8% | estimated ± 8.7 pp, low confidence |
| Vals GPQA Diamond | 91.2% | estimated ± 7.5 pp, medium confidence |
| Vals MMLU-Pro | 84.2% | estimated ± 2.2 pp, high confidence |
| Vals SWE-bench | 82.1% | estimated ± 6.3 pp, low confidence |
| Terminal-Bench 2.1 (Vals AI) | 75.6% | estimated ± 8.5 pp, medium confidence |
| AA AutomationBench | 59.9% | measured |
| GDP.pdf | 18.6% | measured |
| AA Tau3 Banking | 46.9% | estimated ± 5.8 pp, medium confidence |
| AA Terminal-Bench 2.1 | 86.5% | estimated ± 2.6 pp, high confidence |
| AA Terminal-Bench 4.0 | 26.8% | measured |
| CursorBench 4.0 | 37.3% | estimated ± 3.5 pp, high confidence |
| CWE-bench v1 | 56.2% | estimated ± 6.0 pp, medium confidence |
| HLE w/o tools | 32.6% | estimated ± 5.7 pp, medium confidence |
| OSWorld 2.0 | 55.2% | estimated ± 13.3 pp, low confidence |
| PostTrainBench v1.1 | 30.4% | estimated ± 7.2 pp, low confidence |
| Terminal-Bench 4.0 | 30.5% | estimated ± 5.1 pp, medium confidence |
| Terminal-Bench-Science 0.1 | 4.7% | estimated ± 4.1 pp, low confidence |
| Toolathlon-Verified | 76.1% | estimated ± 2.2 pp, medium confidence |
| Vibe Code Bench | 78.4% | measured |
| CharXiv | 83.8% | estimated ± 7.5 pp, low confidence |
| CharXiv w/o tools | 80.5% | estimated ± 2.4 pp, high confidence |
| ExploitGym | 15.0% | estimated ± 5.0 pp, low confidence |
| C-Eval | 95.5% | estimated ± 1.0 pp, low confidence |
| MathVision | 89.1% | estimated ± 4.9 pp, high confidence |
| MCP Atlas | 85.5% | estimated ± 2.7 pp, low confidence |
| MMLU-Redux | 94.7% | estimated ± 1.1 pp, medium confidence |
| MMMU-Pro | 77.6% | estimated ± 4.4 pp, medium confidence |
| NL2Repo | 56.5% | estimated ± 6.5 pp, medium confidence |
| QwenClawBench | 100.0% | estimated ± 6.0 pp, low confidence |
| ScreenSpot Pro | 75.2% | estimated ± 10.7 pp, low confidence |
| SuperGPQA | 70.8% | estimated ± 7.5 pp, medium confidence |
| Toolathlon | 55.9% | estimated ± 6.3 pp, medium confidence |
| ERQA | 65.9% | estimated ± 3.9 pp, high confidence |
| HealthBench Hard | 32.9% | estimated ± 7.7 pp, low confidence |
| MedXpertQA (MM) | 74.3% | estimated ± 6.5 pp, low confidence |
| MMLU-Pro (Arcee) | 100.0% | estimated ± 4.0 pp, low confidence |
| React Native Evals | 85.4% | estimated ± 3.0 pp, low confidence |
| ResearchClawBench | 19.2% | estimated ± 2.2 pp, medium confidence |
| SWE-Rebench | 62.2% | estimated ± 6.5 pp, low confidence |
| ARC-AGI-3 | 1.7% | estimated ± 3.4 pp, medium confidence |
| BioMysteryBench (human-solvable) | 83.8% | estimated ± 0.6 pp, low confidence |
| DRACO | 80.5% | estimated ± 3.0 pp, medium confidence |
| HealthBench (raw) | 44.3% | estimated ± 6.4 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.7 pp, medium confidence |
| HealthBench Professional | 57.7% | estimated ± 4.8 pp, low confidence |
| HealthBench Professional (raw) | 55.0% | estimated ± 5.0 pp, low confidence |
| HLE-Verified | 20.3% | estimated ± 6.0 pp, low confidence |
| LABBench2 | 77.7% | estimated ± 2.4 pp, low confidence |
| MMMU | 84.2% | estimated ± 0.9 pp, high confidence |
| Agents' Last Exam | 39.0% | estimated ± 6.3 pp, medium confidence |
| OpenHarmony Bench | 58.8% | estimated ± 2.4 pp, high confidence |
| MMLU | 89.3% | estimated ± 1.1 pp, medium confidence |
| MMMU-Pro w/ Python | 79.5% | estimated ± 1.6 pp, high confidence |
| RealWorldQA | 85.7% | estimated ± 1.2 pp, medium confidence |
| OmniDocBench 1.5 | 91.7% | estimated ± 6.4 pp, low confidence |