OpenAI · model
o4-mini benchmark scores
As of 2026-10-29, o4-mini (OpenAI) has measured scores on 14 benchmarks and estimated scores on 54 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 78.4% | measured |
| AA-HLE | 16.5% | measured |
| CritPt | 0.6% | measured |
| GDPval-AA | 2.1% | estimated ± 10.9 pp, low confidence |
| GPQA Diamond | 77.8% | estimated ± 1.2 pp, high confidence |
| HLE | 14.4% | estimated ± 6.9 pp, medium confidence |
| IFBench | 66.7% | estimated ± 6.6 pp, medium confidence |
| AA-LCR | 61.0% | measured |
| IFEval | 92.7% | estimated ± 2.8 pp, high confidence |
| LongBench v2 | 58.6% | estimated ± 5.8 pp, low confidence |
| τ²-bench results | 55.6% | measured |
| AA-MMMU-Pro | 69.2% | measured |
| BioMysteryBench (human-difficult) | 11.6% | estimated ± 6.8 pp, low confidence |
| BFCL v4 | 47.1% | estimated ± 8.5 pp, low confidence |
| AA Agentic Index | 11.0% | estimated ± 12.3 pp, low confidence |
| MMLU-Pro | 80.6% | estimated ± 4.0 pp, high confidence |
| AA-IFBench | 68.7% | measured |
| AIME 2024 | 78.8% | estimated ± 8.2 pp, medium confidence |
| Artificial Analysis Intelligence Index | 16.7% | measured |
| BBH | 95.6% | estimated ± 2.4 pp, low confidence |
| GPQA | 77.6% | measured |
| MMLU | 90.0% | measured |
| AA-Omniscience Accuracy | 24.8% | measured |
| MATH-500 | 97.5% | measured |
| MMMLU | 85.9% | estimated ± 0.7 pp, medium confidence |
| MMMU | 79.6% | estimated ± 2.7 pp, high confidence |
| AA MMLU-Pro | 84.8% | estimated ± 0.6 pp, high confidence |
| AIME 2025 | 77.7% | estimated ± 14.2 pp, low confidence |
| ARC-AGI-1 | 40.2% | estimated ± 7.0 pp, low confidence |
| ARC-AGI-2 | 0.0% | estimated ± 11.6 pp, low confidence |
| OfficeQA Pro | 50.3% | estimated ± 6.8 pp, low confidence |
| ParseBench | 53.9% | estimated ± 5.5 pp, medium confidence |
| Vals GPQA Diamond | 66.7% | estimated ± 4.8 pp, medium confidence |
| Vals MMLU-Pro | 59.8% | estimated ± 1.2 pp, medium confidence |
| HLE w/o tools | 31.1% | estimated ± 4.6 pp, high confidence |
| Vibe Code Bench | 20.2% | estimated ± 6.6 pp, low confidence |
| ARC-AGI-3 | 1.4% | estimated ± 7.7 pp, low confidence |
| CharXiv | 76.8% | estimated ± 6.2 pp, low confidence |
| CharXiv w/o tools | 76.1% | estimated ± 2.6 pp, medium confidence |
| AI2D_TEST | 88.1% | estimated ± 2.0 pp, medium confidence |
| C-Eval | 90.6% | estimated ± 1.0 pp, medium confidence |
| CountBench | 90.9% | estimated ± 2.9 pp, low confidence |
| DynaMath | 59.5% | estimated ± 3.0 pp, low confidence |
| ERQA | 40.9% | estimated ± 2.9 pp, medium confidence |
| MathVision | 76.7% | estimated ± 3.2 pp, medium confidence |
| MMLU-Redux | 94.1% | estimated ± 0.8 pp, medium confidence |
| MMMU-Pro | 70.7% | estimated ± 1.6 pp, high confidence |
| RealWorldQA | 79.1% | estimated ± 2.7 pp, medium confidence |
| ScreenSpot Pro | 70.3% | estimated ± 11.2 pp, low confidence |
| SuperGPQA | 58.5% | estimated ± 8.2 pp, low confidence |
| V* | 39.3% | estimated ± 7.5 pp, low confidence |
| Video-MME (w/o subtitle) | 79.2% | estimated ± 1.7 pp, medium confidence |
| Video-MME (with subtitle) | 81.2% | estimated ± 3.7 pp, low confidence |
| AA LiveCodeBench | 85.9% | measured |
| HealthBench Hard | 6.5% | estimated ± 8.6 pp, low confidence |
| MedXpertQA (MM) | 45.3% | estimated ± 6.5 pp, low confidence |
| MMLU-Pro (Arcee) | 79.4% | estimated ± 4.0 pp, medium confidence |
| BioMysteryBench (human-solvable) | 81.7% | estimated ± 0.6 pp, low confidence |
| HealthBench (raw) | 23.4% | estimated ± 6.0 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.3 pp, medium confidence |
| HealthBench Professional | 58.1% | estimated ± 4.5 pp, low confidence |
| HealthBench Professional (raw) | 35.8% | estimated ± 4.5 pp, medium confidence |
| HLE-Verified | 19.7% | estimated ± 7.1 pp, low confidence |
| STEMBench | 2.4% | estimated ± 0.2 pp, medium confidence |
| SimpleQA | 20.2% | measured |
| MRCRv2 | 28.9% | estimated ± 11.3 pp, low confidence |
| MMLongBench-Doc | 60.4% | estimated ± 2.6 pp, low confidence |
| MMMU-Pro w/ Python | 73.0% | estimated ± 2.0 pp, high confidence |