DeepSeek · model
DeepSeek-R1-0528 benchmark scores
As of 2026-10-29, DeepSeek-R1-0528 (DeepSeek) has measured scores on 9 benchmarks and estimated scores on 30 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 80.9% | estimated ± 3.0 pp, medium confidence |
| AA-HLE | 15.8% | estimated ± 5.0 pp, high confidence |
| AA-SciCode | 37.3% | estimated ± 3.9 pp, high confidence |
| AIME26 | 87.5% | estimated ± 4.6 pp, medium confidence |
| CritPt | 2.2% | estimated ± 3.4 pp, medium confidence |
| GPQA Diamond | 81.1% | estimated ± 1.2 pp, high confidence |
| HLE | 17.7% | measured |
| SciCode | 37.1% | estimated ± 3.3 pp, high confidence |
| AA Coding Index | 26.8% | estimated ± 5.3 pp, medium confidence |
| SWE-bench Verified | 57.6% | measured |
| SWE-bench Pro | 33.6% | estimated ± 4.6 pp, high confidence |
| MMLU-Pro | 85.0% | measured |
| AIME 2024 | 91.4% | measured |
| Artificial Analysis Intelligence Index | 21.9% | estimated ± 6.3 pp, low confidence |
| BBH | 98.8% | estimated ± 2.8 pp, medium confidence |
| GPQA | 81.0% | measured |
| MMLU | 90.6% | estimated ± 2.2 pp, medium confidence |
| AA-Omniscience Accuracy | 26.5% | estimated ± 2.7 pp, medium confidence |
| MATH-500 | 98.0% | estimated ± 1.2 pp, medium confidence |
| MMMLU | 86.1% | estimated ± 1.4 pp, medium confidence |
| AA MMLU-Pro | 86.0% | estimated ± 1.0 pp, low confidence |
| AIME 2025 | 87.5% | measured |
| ARC-AGI-1 | 51.8% | estimated ± 7.0 pp, low confidence |
| ARC-AGI-2 | 0.0% | estimated ± 11.6 pp, low confidence |
| Vals GPQA Diamond | 72.0% | estimated ± 4.8 pp, medium confidence |
| Vals LiveCodeBench | 39.4% | estimated ± 9.9 pp, low confidence |
| Vals MMLU-Pro | 72.7% | estimated ± 1.2 pp, medium confidence |
| Vals SWE-bench | 45.6% | estimated ± 6.8 pp, low confidence |
| HLE w/o tools | 14.1% | estimated ± 5.2 pp, medium confidence |
| Vibe Code Bench | 0.6% | estimated ± 8.1 pp, low confidence |
| C-Eval | 88.1% | estimated ± 0.9 pp, low confidence |
| HMMT Feb 2025 | 84.6% | estimated ± 3.1 pp, high confidence |
| MMLU-Redux | 93.4% | measured |
| NL2Repo | 9.6% | estimated ± 6.2 pp, low confidence |
| SuperGPQA | 66.7% | estimated ± 4.1 pp, high confidence |
| React Native Evals | 71.4% | estimated ± 2.3 pp, low confidence |
| SWE-Rebench | 57.0% | estimated ± 3.6 pp, low confidence |
| SimpleQA | 27.8% | measured |
| LiveCodeBench | 73.3% | measured |