DeepSeek · model
DeepSeek V3.2 benchmark scores
As of 2026-11-22, DeepSeek V3.2 (DeepSeek) has measured scores on 16 benchmarks and estimated scores on 77 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 75.1% | measured |
| AA-HLE | 11.2% | measured |
| AA-SciCode | 45.4% | estimated ± 8.6 pp, low confidence |
| Artificial Analysis Intelligence Index | 16.0% | measured |
| CritPt | 0.9% | measured |
| GDPval-AA | 15.9% | estimated ± 10.5 pp, low confidence |
| GPQA Diamond | 74.5% | estimated ± 2.4 pp, high confidence |
| HLE | 18.8% | estimated ± 8.5 pp, medium confidence |
| IFBench | 55.6% | estimated ± 6.6 pp, medium confidence |
| AA-LCR | 45.7% | measured |
| AA-Omniscience Accuracy | 24.0% | measured |
| BrowseComp | 50.9% | estimated ± 8.0 pp, low confidence |
| HLE w/ tools | 30.5% | estimated ± 5.5 pp, low confidence |
| IFEval | 90.9% | estimated ± 2.8 pp, high confidence |
| LongBench v2 | 51.8% | estimated ± 5.5 pp, medium confidence |
| VITA-Bench | 18.5% | measured |
| τ²-bench results | 78.9% | measured |
| AA Coding Index | 40.1% | estimated ± 11.1 pp, low confidence |
| APEX-Agents-AA | 9.6% | estimated ± 9.3 pp, medium confidence |
| BioMysteryBench (human-difficult) | 15.7% | estimated ± 6.8 pp, low confidence |
| SWE-bench Verified | 78.0% | estimated ± 4.8 pp, medium confidence |
| BFCL v4 | 60.8% | estimated ± 8.5 pp, low confidence |
| CyberGym | 42.2% | estimated ± 11.0 pp, low confidence |
| JobBench | 0.0% | estimated ± 9.3 pp, low confidence |
| WideResearch | 63.1% | estimated ± 3.3 pp, medium confidence |
| AA Agentic Index | 12.5% | estimated ± 9.4 pp, medium confidence |
| MMLU-Pro | 84.9% | estimated ± 5.0 pp, medium confidence |
| AA-IFBench | 49.0% | measured |
| FailureSensorIQ | 62.0% | estimated ± 1.7 pp, medium confidence |
| GPQA | 77.0% | estimated ± 6.1 pp, medium confidence |
| MMLU | 90.4% | estimated ± 0.9 pp, high confidence |
| Vals GPQA Diamond | 74.2% | estimated ± 3.2 pp, high confidence |
| Vals MMLU-Pro | 82.0% | estimated ± 2.9 pp, high confidence |
| MMMLU | 86.3% | estimated ± 0.7 pp, medium confidence |
| Chess Puzzles | 19.3% | estimated ± 12.3 pp, low confidence |
| AA MMLU-Pro | 84.7% | estimated ± 0.8 pp, high confidence |
| Gert Labs | 21.1% | measured |
| ApprenticeBench | 3.1% | estimated ± 10.4 pp, low confidence |
| ARC-AGI-1 | 75.4% | estimated ± 7.8 pp, low confidence |
| ARC-AGI-2 | 32.3% | estimated ± 14.4 pp, low confidence |
| MazeBench | 0.0% | estimated ± 2.7 pp, medium confidence |
| Mystery Game Puzzles | 14.1% | estimated ± 2.2 pp, low confidence |
| NYT Connections | 8.2% | measured |
| OSWorld-Verified | 38.7% | estimated ± 3.8 pp, low confidence |
| Vals SWE-bench | 63.5% | estimated ± 7.0 pp, low confidence |
| CWE-Bench-AA | 32.3% | estimated ± 8.1 pp, low confidence |
| AA Cyber Index | 4.3% | estimated ± 12.4 pp, low confidence |
| DeepsecBench-AA | 6.2% | estimated ± 8.2 pp, low confidence |
| HealthBench Professional (length-adjusted, Artificial Analysis) | 50.7% | estimated ± 4.3 pp, low confidence |
| HealthBench Professional (raw, Artificial Analysis) | 51.2% | estimated ± 1.1 pp, low confidence |
| AA ITBench | 35.7% | estimated ± 4.6 pp, low confidence |
| CWE-bench v1 | 33.7% | estimated ± 5.7 pp, low confidence |
| HLE w/o tools | 28.7% | estimated ± 4.9 pp, high confidence |
| OSWorld 2.1 | 1.6% | estimated ± 3.6 pp, medium confidence |
| StealthMark | 41.3% | estimated ± 10.7 pp, low confidence |
| Toolathlon-Verified | 55.8% | estimated ± 3.0 pp, high confidence |
| AA AutomationBench | 38.4% | estimated ± 9.9 pp, low confidence |
| GDP.pdf | 10.0% | estimated ± 4.2 pp, medium confidence |
| ARC-AGI-3 | 1.3% | estimated ± 9.8 pp, low confidence |
| ARC-AGI-3 Provider Adapter | 2.7% | estimated ± 7.3 pp, low confidence |
| CWE-bench | 26.5% | estimated ± 6.2 pp, low confidence |
| MKA | 32.8% | estimated ± 4.6 pp, medium confidence |
| SkillsBench official leaderboard (BenchFlow) | 19.2% | estimated ± 6.1 pp, medium confidence |
| BinaryAudit | 12.3% | measured |
| OTelBench | 11.6% | measured |
| Vibe Code Bench | 27.7% | estimated ± 6.5 pp, low confidence |
| Hermes Index | 15.0% | estimated ± 5.7 pp, low confidence |
| ExploitGym | 0.0% | estimated ± 8.3 pp, low confidence |
| C-Eval | 90.4% | estimated ± 1.0 pp, medium confidence |
| Claw-Eval | 40.2% | measured |
| DeepPlanning | 26.2% | estimated ± 9.1 pp, low confidence |
| MCP Atlas | 40.9% | estimated ± 13.3 pp, low confidence |
| MMLU-Redux | 94.0% | estimated ± 0.8 pp, medium confidence |
| QwenClawBench | 52.0% | estimated ± 2.0 pp, low confidence |
| SuperGPQA | 65.5% | estimated ± 9.3 pp, medium confidence |
| Toolathlon | 37.8% | estimated ± 8.0 pp, medium confidence |
| HealthBench Hard | 8.9% | estimated ± 8.6 pp, low confidence |
| MMLU-Pro (Arcee) | 80.5% | estimated ± 4.0 pp, medium confidence |
| React Native Evals | 71.5% | measured |
| ResearchClawBench | 11.3% | estimated ± 1.3 pp, medium confidence |
| SWE-Rebench | 60.9% | measured |
| Finance Agent v2 | 55.1% | estimated ± 2.4 pp, low confidence |
| BioMysteryBench (human-solvable) | 81.7% | estimated ± 0.6 pp, low confidence |
| HealthBench (length-adjusted) | 54.1% | estimated ± 3.7 pp, medium confidence |
| HealthBench Professional | 56.7% | estimated ± 4.2 pp, low confidence |
| HLE-Verified | 18.9% | estimated ± 7.1 pp, low confidence |
| STEMBench | 2.4% | estimated ± 0.2 pp, low confidence |
| SimpleQA | 23.6% | estimated ± 4.3 pp, medium confidence |
| Agents' Last Exam | 29.6% | estimated ± 9.7 pp, low confidence |
| MRCRv2 | 22.8% | estimated ± 11.3 pp, low confidence |
| MMLongBench-Doc | 59.0% | estimated ± 2.6 pp, low confidence |
| BioBasicsBench | 34.1% | estimated ± 3.8 pp, low confidence |
| LifeVal | 0.9% | estimated ± 5.0 pp, low confidence |