Mistral · model
Mistral Medium 3.5 128B benchmark scores
As of 2026-10-07, Mistral Medium 3.5 128B (Mistral) has measured scores on 20 benchmarks and estimated scores on 91 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 74.8% | measured |
| AA-HLE | 13.8% | measured |
| AA-SciCode | 40.2% | measured |
| Artificial Analysis Intelligence Index | 14.2% | measured |
| CritPt | 0.0% | measured |
| GDPval-AA | 13.2% | measured |
| GPQA Diamond | 48.6% | estimated ± 1.7 pp, medium confidence |
| HLE | 19.0% | estimated ± 9.5 pp, medium confidence |
| IFBench | 67.6% | estimated ± 6.9 pp, medium confidence |
| AA-LCR | 69.3% | measured |
| LiveCodeBench v6 | 84.3% | estimated ± 3.1 pp, high confidence |
| AA-Omniscience Accuracy | 24.7% | measured |
| SciCode | 40.2% | estimated ± 3.2 pp, high confidence |
| Terminal-Bench 2.1 | 30.6% | estimated ± 6.9 pp, low confidence |
| BrowseComp | 58.7% | estimated ± 5.7 pp, medium confidence |
| HLE w/ tools | 38.4% | estimated ± 6.1 pp, low confidence |
| IFEval | 93.3% | estimated ± 1.3 pp, high confidence |
| LongBench v2 | 61.5% | estimated ± 5.9 pp, medium confidence |
| τ²-bench results | 94.2% | measured |
| AA Coding Index | 46.9% | measured |
| AA-MMMU-Pro | 64.9% | measured |
| APEX-Agents | 12.8% | estimated ± 10.0 pp, low confidence |
| APEX-Agents-AA | 13.2% | estimated ± 9.1 pp, medium confidence |
| BioMysteryBench (human-difficult) | 31.1% | estimated ± 5.9 pp, low confidence |
| DeepSearchQA | 71.1% | estimated ± 4.5 pp, medium confidence |
| SWE-bench Verified | 77.6% | measured |
| AutomationBench | 28.4% | estimated ± 9.7 pp, low confidence |
| BFCL v4 | 70.0% | estimated ± 2.6 pp, high confidence |
| CyberGym | 52.7% | estimated ± 5.3 pp, low confidence |
| JobBench | 6.4% | estimated ± 8.6 pp, low confidence |
| SWE-bench Pro | 53.9% | estimated ± 3.2 pp, high confidence |
| τ³-bench results | 91.4% | measured |
| WideResearch | 56.0% | estimated ± 5.9 pp, low confidence |
| AA Agentic Index | 9.4% | measured |
| MMLU-Pro | 85.5% | estimated ± 3.5 pp, high confidence |
| GPQA | 79.6% | estimated ± 1.6 pp, medium confidence |
| AA-IFBench | 68.8% | measured |
| Gert Labs | 39.1% | measured |
| AA-AnalystAgent | 8.6% | estimated ± 12.0 pp, low confidence |
| AA EnterpriseOps-Gym | 45.1% | estimated ± 3.6 pp, medium confidence |
| AA Harvey LAB | 79.4% | estimated ± 1.2 pp, low confidence |
| ApprenticeBench | 0.0% | estimated ± 3.1 pp, low confidence |
| ARC-AGI-1 | 14.1% | estimated ± 5.4 pp, low confidence |
| ARC-AGI-2 | 0.0% | estimated ± 12.6 pp, low confidence |
| CursorBench 3.1 | 19.8% | estimated ± 3.4 pp, low confidence |
| CursorBench 3.2 | 32.0% | estimated ± 4.4 pp, medium confidence |
| Terminal-Bench 3.0 | 0.4% | estimated ± 6.9 pp, low confidence |
| FrontierCode 1.1 Main | 17.4% | estimated ± 2.7 pp, low confidence |
| FrontierSWE v2 | 5.5% | estimated ± 9.5 pp, low confidence |
| OfficeQA Pro | 50.3% | estimated ± 7.2 pp, low confidence |
| OSWorld-Verified | 54.0% | estimated ± 5.9 pp, medium confidence |
| Vals GPQA Diamond | 34.8% | measured |
| Vals LiveCodeBench | 79.6% | estimated ± 9.9 pp, low confidence |
| Vals MMLU-Pro | 75.3% | measured |
| Vals SWE-bench | 66.4% | measured |
| Terminal-Bench 2.1 (Vals AI) | 39.0% | measured |
| AA AutomationBench | 3.8% | estimated ± 7.0 pp, low confidence |
| GDP.pdf | 9.4% | estimated ± 5.4 pp, low confidence |
| AA ITBench | 46.1% | estimated ± 3.6 pp, medium confidence |
| AA Tau3 Banking | 22.0% | estimated ± 5.8 pp, low confidence |
| AA Terminal-Bench 2.1 | 26.3% | estimated ± 4.4 pp, medium confidence |
| AA Terminal-Bench 4.0 | 0.0% | estimated ± 14.0 pp, low confidence |
| CursorBench 4.0 | 14.3% | estimated ± 1.6 pp, medium confidence |
| CWE-bench v1 | 20.9% | estimated ± 6.9 pp, low confidence |
| DeepSWE | 10.1% | estimated ± 6.4 pp, low confidence |
| HLE w/o tools | 25.1% | estimated ± 5.1 pp, low confidence |
| OSWorld 2.0 | 0.0% | estimated ± 4.5 pp, low confidence |
| PostTrainBench v1.1 | 18.9% | estimated ± 3.3 pp, medium confidence |
| Toolathlon-Verified | 28.9% | estimated ± 2.3 pp, low confidence |
| Vibe Code Bench | 18.6% | estimated ± 8.1 pp, medium confidence |
| CharXiv | 70.4% | estimated ± 7.5 pp, low confidence |
| CharXiv w/o tools | 72.8% | estimated ± 2.4 pp, medium confidence |
| ExploitGym | 0.0% | estimated ± 12.2 pp, low confidence |
| C-Eval | 87.8% | estimated ± 1.0 pp, low confidence |
| Claw-Eval | 54.8% | estimated ± 6.8 pp, low confidence |
| DeepPlanning | 35.4% | estimated ± 8.3 pp, low confidence |
| MathVision | 81.5% | estimated ± 4.9 pp, high confidence |
| MCP Atlas | 64.6% | estimated ± 7.8 pp, low confidence |
| MCP-Tasks | 100.0% | estimated ± 3.6 pp, low confidence |
| MMLU-Redux | 94.5% | estimated ± 1.1 pp, medium confidence |
| MMMU-Pro | 72.2% | estimated ± 4.4 pp, medium confidence |
| NL2Repo | 39.5% | estimated ± 4.7 pp, medium confidence |
| QwenClawBench | 55.9% | estimated ± 6.0 pp, low confidence |
| ScreenSpot Pro | 64.7% | estimated ± 10.7 pp, low confidence |
| SuperGPQA | 69.7% | estimated ± 7.5 pp, medium confidence |
| Terminal-Bench 2.0 | 49.0% | estimated ± 5.9 pp, medium confidence |
| Toolathlon | 83.3% | estimated ± 6.0 pp, low confidence |
| ERQA | 4.2% | estimated ± 3.9 pp, medium confidence |
| HealthBench Hard | 1.2% | estimated ± 7.7 pp, low confidence |
| MedXpertQA (MM) | 16.9% | estimated ± 6.5 pp, low confidence |
| MMLU-Pro (Arcee) | 77.1% | estimated ± 4.0 pp, low confidence |
| React Native Evals | 60.6% | estimated ± 2.4 pp, low confidence |
| ResearchClawBench | 11.6% | estimated ± 2.0 pp, medium confidence |
| SWE-Rebench | 60.9% | estimated ± 3.6 pp, medium confidence |
| ARC-AGI-3 | 1.7% | estimated ± 3.4 pp, medium confidence |
| BioMysteryBench (human-solvable) | 81.7% | estimated ± 0.6 pp, low confidence |
| DRACO | 67.0% | estimated ± 3.0 pp, low confidence |
| FrontierCode 1.1 Extended | 12.0% | estimated ± 2.4 pp, low confidence |
| HealthBench (raw) | 20.2% | estimated ± 6.4 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.7 pp, low confidence |
| HealthBench Professional | 49.8% | estimated ± 1.3 pp, low confidence |
| HealthBench Professional (raw) | 33.0% | estimated ± 5.0 pp, low confidence |
| HLE-Verified | 15.3% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 74.6% | estimated ± 1.8 pp, low confidence |
| MMMU | 75.8% | estimated ± 0.9 pp, high confidence |
| LiveCodeBench | 90.8% | estimated ± 8.9 pp, low confidence |
| OpenHarmony Bench | 51.0% | estimated ± 2.4 pp, medium confidence |
| MMLU | 88.8% | estimated ± 1.1 pp, medium confidence |
| MMMU-Pro w/ Python | 71.9% | estimated ± 1.6 pp, medium confidence |
| RealWorldQA | 79.0% | estimated ± 1.2 pp, low confidence |
| OmniDocBench 1.5 | 0.0% | estimated ± 6.4 pp, low confidence |