Other · model
MAI-Thinking-1 benchmark scores
As of 2026-11-21, MAI-Thinking-1 (Other) has measured scores on 12 benchmarks and estimated scores on 71 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 82.9% | estimated ± 4.4 pp, high confidence |
| AA-HLE | 23.6% | estimated ± 3.9 pp, high confidence |
| AA-SciCode | 46.6% | estimated ± 3.6 pp, high confidence |
| AIME26 | 94.5% | measured |
| Artificial Analysis Intelligence Index | 17.5% | estimated ± 7.4 pp, medium confidence |
| CritPt | 1.3% | estimated ± 4.0 pp, high confidence |
| GPQA Diamond | 84.2% | measured |
| HLE | 22.8% | estimated ± 6.5 pp, medium confidence |
| IFBench | 85.0% | measured |
| LiveCodeBench v6 | 87.7% | measured |
| AA-Omniscience Accuracy | 21.9% | estimated ± 8.4 pp, medium confidence |
| SciCode | 44.6% | estimated ± 3.6 pp, high confidence |
| IFEval | 95.3% | estimated ± 3.6 pp, high confidence |
| AA Coding Index | 47.2% | estimated ± 5.3 pp, medium confidence |
| SWE-bench Verified | 73.5% | measured |
| SWE-bench Pro | 52.8% | measured |
| MMLU-Pro | 85.0% | measured |
| AA-IFBench | 80.4% | estimated ± 7.4 pp, low confidence |
| AIME 2024 | 95.5% | estimated ± 2.6 pp, high confidence |
| BBH | 98.8% | estimated ± 2.8 pp, medium confidence |
| GPQA | 84.2% | measured |
| MMLU | 90.6% | estimated ± 2.1 pp, medium confidence |
| Vals GPQA Diamond | 83.6% | estimated ± 2.1 pp, medium confidence |
| Vals LiveCodeBench | 86.1% | estimated ± 3.4 pp, medium confidence |
| Vals MMLU-Pro | 83.4% | estimated ± 1.6 pp, medium confidence |
| MATH-500 | 99.1% | estimated ± 2.0 pp, medium confidence |
| MMMLU | 87.3% | estimated ± 1.1 pp, medium confidence |
| Chess Puzzles | 16.0% | estimated ± 10.8 pp, low confidence |
| AIME 2025 | 97.0% | measured |
| AA MMLU-Pro | 86.4% | estimated ± 1.2 pp, low confidence |
| ARC-AGI-1 | 79.3% | estimated ± 2.5 pp, medium confidence |
| ARC-AGI-2 | 25.3% | estimated ± 9.7 pp, low confidence |
| CursorBench 3.1 | 49.7% | estimated ± 5.6 pp, low confidence |
| CursorBench 3.2 | 46.2% | estimated ± 3.6 pp, medium confidence |
| EBR-bench | 9.5% | estimated ± 11.3 pp, low confidence |
| FrontierCode 1.1 Main | 16.0% | estimated ± 4.2 pp, low confidence |
| FrontierMath v2 (Tier 4) | 5.3% | estimated ± 4.1 pp, low confidence |
| FrontierMath v2 (Tiers 1-3) | 34.2% | estimated ± 8.8 pp, medium confidence |
| FrontierSWE v2 | 3.1% | estimated ± 8.3 pp, low confidence |
| Furniture Assembly | 33.3% | estimated ± 12.4 pp, low confidence |
| MaxxBench | 32.9% | estimated ± 5.7 pp, low confidence |
| Mystery Game Puzzles | 7.1% | estimated ± 13.3 pp, low confidence |
| Terminal-Bench Hard | 41.8% | estimated ± 4.0 pp, medium confidence |
| Vals SWE-bench | 67.4% | estimated ± 6.6 pp, medium confidence |
| Terminal-Bench 2.1 (Vals AI) | 40.1% | estimated ± 6.4 pp, medium confidence |
| Bug Hunt Bench | 28.0% | estimated ± 12.1 pp, low confidence |
| CursorBench 4.0 | 30.1% | estimated ± 3.3 pp, medium confidence |
| DeepSWE | 47.7% | estimated ± 11.2 pp, low confidence |
| FrontierCode 2.0 | 19.4% | estimated ± 6.7 pp, low confidence |
| FrontierPhysics | 1.0% | estimated ± 4.7 pp, medium confidence |
| HLE w/o tools | 20.3% | estimated ± 4.3 pp, high confidence |
| PostTrainBench v1.1 | 21.5% | estimated ± 4.4 pp, medium confidence |
| StealthMark | 14.1% | measured |
| VulcanBench Frontier v4 | 75.4% | estimated ± 8.3 pp, low confidence |
| ARC-AGI-3 | 0.0% | estimated ± 8.6 pp, low confidence |
| Vibe Code Bench | 0.7% | estimated ± 8.4 pp, medium confidence |
| FACTS Multimodal | 39.3% | estimated ± 4.7 pp, medium confidence |
| ExploitBench | 5.6% | estimated ± 12.6 pp, low confidence |
| C-Eval | 91.7% | estimated ± 1.2 pp, high confidence |
| HMMT Feb 2025 | 93.9% | estimated ± 1.8 pp, low confidence |
| HMMT Feb 2026 | 84.9% | measured |
| HMMT Nov 2025 | 92.3% | estimated ± 2.8 pp, high confidence |
| MMAnswerBench | 81.9% | estimated ± 1.3 pp, high confidence |
| MMLU-Redux | 93.2% | estimated ± 0.6 pp, high confidence |
| NL2Repo | 35.4% | estimated ± 2.8 pp, high confidence |
| PostTrainBench v1 | 17.7% | estimated ± 4.6 pp, medium confidence |
| SuperGPQA | 67.1% | estimated ± 8.5 pp, low confidence |
| Terminal-Bench 2.0 | 46.0% | measured |
| LiveCodeBench Pro | 77.8% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (Text) | 40.1% | estimated ± 4.5 pp, low confidence |
| MMLU-Pro (Arcee) | 84.8% | estimated ± 3.4 pp, medium confidence |
| React Native Evals | 76.7% | estimated ± 2.7 pp, medium confidence |
| SWE-Rebench | 57.0% | estimated ± 3.6 pp, medium confidence |
| PostTrainBench | 19.8% | estimated ± 3.7 pp, medium confidence |
| DAYJOB | 0.0% | estimated ± 3.2 pp, medium confidence |
| DAYJOB Healthcare | 0.0% | estimated ± 4.7 pp, medium confidence |
| FrontierCode 1.1 Extended | 53.4% | estimated ± 3.0 pp, low confidence |
| SimpleQA | 39.8% | estimated ± 11.7 pp, low confidence |
| LiveCodeBench | 75.3% | estimated ± 14.6 pp, low confidence |
| OpenHarmony Bench v1.0 | 53.3% | estimated ± 4.9 pp, low confidence |
| SEC-Bench Pro | 13.9% | estimated ± 9.3 pp, low confidence |
| IMOAnswerBench | 80.4% | estimated ± 5.6 pp, medium confidence |
| ExploitBench Internal Port | 0.0% | estimated ± 6.8 pp, low confidence |