OpenAI
gpt-oss-20b benchmark results
9 results on 5 benchmarks from 3 sources.
Also listed as: gpt-oss-20b_medium · gpt-oss-20b_low · gpt-oss-20b_high · gpt-oss-20b_unknown · GPT-OSS-20B · GPT-oss-20B(medium) · GPT-oss-20B(high)
All benchmark results
| Benchmarks | Score | Rank | Setting | Source type | Date |
|---|---|---|---|---|---|
| LMArena Text | 1317 | 265 | Overall, style control on; 10418 votes; 95% CI 1311–1324 | Official leaderboard | Oct 8, 2026 |
| GPQA Diamond | 60.8% | 182 | Epoch AI run, accuracy | Independent run | Aug 27, 2026 |
| MMLU-Pro | 73.6% | 89 | Overall accuracy; data source: Self-Reported; 20B“GPT-oss-20B(high)” | Vendor-reported | Mar 11, 2026 |
| MMLU-Pro | 73.1% | 91 | Overall accuracy; data source: TIGER-Lab; 20B“GPT-oss-20B(medium)” | Official leaderboard | Mar 11, 2026 |
| AIME (OTIS Mock AIME 2024–2025) | 50.8% | 167 | Epoch AI run, OTIS Mock AIME 2024–2025 | Independent run | Aug 27, 2026 |
| Terminal-Bench 2.0 | 3.4% | 194 | Terminal-Bench 2.0; agent: Mini-SWE-Agent | Official leaderboard via Epoch AI | Nov 3, 2025 |
| Terminal-Bench 2.0 | 3.4% | 195 | Terminal-Bench 2.0; agent: Mini-SWE-Agent | Official leaderboard via Epoch AI | Aug 5, 2025 |
| Terminal-Bench 2.0 | 3.1% | 196 | Terminal-Bench 2.0; agent: Terminus 2 | Official leaderboard via Epoch AI | Nov 1, 2025 |
| Terminal-Bench 2.0 | 3.1% | 197 | Terminal-Bench 2.0; agent: Terminus 2 | Official leaderboard via Epoch AI | Aug 5, 2025 |
Related on this site
FAQ
Where do these gpt-oss-20b scores come from?
From Epoch AI Benchmarking Hub, LMArena leaderboard dataset, and MMLU-Pro leaderboard (TIGER-Lab). Each row links to the original leaderboard, dataset or report, and vendor-reported numbers are labelled.