gpt-oss-20b places in the top three on AIME 2024.
Open weights
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMLU-Pro | 73.6% | #77 of 212 | Official leaderboard | |
| MMLU | 85.3% | #53 of 127 | Model card | |
| SuperGPQA | 44.9% | #29 of 58 | Official leaderboard | |
| LMArena (Elo) | 1,317 | #212 of 333 | Official leaderboard | |
| Vectara Hallucination Leaderboard (HHEM) | 3.70% | #95 of 148 | Official leaderboard | |
| Wer wird Millionär? LLM Benchmark | 296.0 | #55 of 76 | Official leaderboard | |
| Humanity's Last Exam | 17.3% | #111 of 147 | Model card | |
| MMMLU | 75.7% | #47 of 62 | Model card | |
| SEA-HELM | 60.6 | #48 of 62 | Official leaderboard | |
| EQ-Bench 3 | 985 | #64 of 78 | Official leaderboard | |
| AgentIF (2026-10 setting) | 55% | #21 of 24 | Official leaderboard | |
| SimpleQA | 6.70% | #54 of 59 | Model card | |
| SimpleQA Verified | 5.95% | #53 of 56 | Official leaderboard | |
| Judgemark v2.1 | 42.8 | #40 of 42 | Official leaderboard | |
| MultiNRC | 10.4% | #37 of 38 | Official leaderboard | |
| FACTS Parametric | 2.90% | #44 of 45 | Official leaderboard | |
| Longform Creative Writing (EQ-Bench) | 21.2 | #118 of 119 | Official leaderboard | |
| Creative Writing v3 | 40.5 | #119 of 120 | Official leaderboard | |
| FACTS Search | 34.8% | #31 of 31 | Official leaderboard | |
| IndicGenBench | 34.5 | #13 of 13 | Official leaderboard | |
| SimpleQA Hallucination Rate | 91.4% | #12 of 12 | Model card | |
| PersonQA | 15.5% | #9 of 9 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AIME 2024 | 96% | #2 of 87 | Model card | |
| AIME 2025 | 98.7% | #13 of 143 | Model card | |
| HMMT 2025 | 76.7% | #27 of 53 | Official leaderboard | |
| CritPt | 1.43% | #87 of 160 | Epoch AI | |
| OTIS Mock AIME 2024-2025 | 65.3% | #110 of 202 | Epoch AI | |
| GPQA Diamond | 60.8% | #126 of 219 | Epoch AI | |
| DTBench | 46.7% | #116 of 178 | Official leaderboard | |
| SMT 2025 | 83% | #27 of 40 | Official leaderboard | |
| CMIMC 2025 | 73.8% | #23 of 33 | Official leaderboard | |
| BRUMO 2025 | 86.7% | #30 of 42 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Codeforces rating | 2,516 | #11 of 41 | Model card | |
| WeirdML | 40.9% | #75 of 112 | Official leaderboard | |
| SWE-bench Verified | 60.7% | #105 of 150 | Model card | |
| ALE-Bench | 566 | #97 of 121 | Epoch AI | |
| OpenAI PRs | 37% | #10 of 12 | Model card | |
| Aider Polyglot | 34.2% | #26 of 29 | Model card | |
| PaperBench | 7.00% | #14 of 14 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LisanBench | 569.0 | #65 of 133 | Official leaderboard | |
| BalatroBench | 5.80 | #11 of 16 | Official leaderboard | |
| LLM Chess | -232 | #98 of 124 | Official leaderboard | |
| tau-bench | 54.8% | #19 of 23 | Model card | |
| PinchBench | 48.4% | #43 of 48 | Official leaderboard | |
| PinchBench 2.0 | 36.3% | #54 of 57 | Official leaderboard | |
| Terminal-Bench 2.0 | 3.37% | #43 of 43 | Official leaderboard | |
| Terminal-Bench | 3.37% | #18 of 18 | Epoch AI |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| CorpFin v2 | 53.1% | #91 of 119 | Official leaderboard | |
| ProfBench | 42.3% | #56 of 72 | Official leaderboard | |
| LegalBench | 70.8% | #113 of 138 | Official leaderboard | |
| TaxEval v2 | 63.7% | #108 of 129 | Official leaderboard | |
| LEXam | 32.1% | #26 of 31 | Official leaderboard | |
| FailureSensorIQ | 43.5% | #22 of 24 | Official leaderboard | |
| CaseLaw v2 | 43.8% | #53 of 53 | Official leaderboard | |
| HealthBench | 42.5% | #30 of 30 | Model card | |
| HealthBench Hard | 10.8% | #21 of 21 | Model card | |
| HealthBench Consensus | 82.6% | #17 of 17 | Model card | |
| StockBench | -2.80% | #13 of 13 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AIR-Bench 2024 | 86% | #10 of 82 | Official leaderboard | |
| Anthropic Red Team | 99.7% | #12 of 75 | Official leaderboard | |
| HELM Safety | 96.4 | #16 of 75 | Official leaderboard | |
| SimpleSafetyTests | 100% | #19 of 75 | Official leaderboard | |
| FORTRESS | 17.6% | #21 of 66 | Official leaderboard | |
| Tacit Knowledge and Troubleshooting | 73% | #11 of 29 | Model card | |
| Capture the Flag (Collegiate) | 17% | #8 of 13 | Model card | |
| Multimodal Troubleshooting Virology | 50% | #18 of 26 | Model card | |
| BrowseSafe++ | 71% | #12 of 15 | Official leaderboard | |
| Human Pathogen Capabilities Test v2 (HPCT v2) | 37.9% | #57 of 71 | Official leaderboard | |
| KORA | 16.1% | #38 of 42 | Official leaderboard | |
| Capture the Flag (Professional) | 2.00% | #20 of 21 | Model card | |
| ProtocolQA Open-Ended | 16% | #25 of 25 | Model card | |
| TroubleshootingBench | 19% | #19 of 19 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HELM Capabilities | 67.4 | #27 of 60 | Official leaderboard | |
| ACCoRD | 61.9 | #82 of 150 | Official leaderboard | |
| Conceptual Reasoning Index | 30.4 | #114 of 149 | Official leaderboard | |
| LMCA | 14.5% | #120 of 139 | Epoch AI |
OpenAI's models in color, against every other model on the Epoch Capabilities Index, by release date.
The benchmarks the lab chose to headline, with the scores it gave.
Every result in the longer documents the lab published with it. 21 of them aren't in the launch post.