OpenAIGPT OSS
GPT-OSS-120B
OpenAI · GPT OSS. Released Aug 5, 2025.
GPT-OSS-120B is a model from OpenAI in the GPT OSS family, released Aug 5, 2025. evals.report tracks 25 reported GPT-OSS-120B benchmark scores across LiveCodeBench Pro, SWE-bench Pro, AIME (OTIS Mock), GPQA Diamond, SimpleQA Verified, LMArena, WeirdML, Artificial Analysis Intelligence Index, and 17 more — each shown with its benchmark, metric, source status, and date, and never combined into a single ranking.
Open25 results
Benchmark results 25
Compare this model| Benchmark | Category | Score | Metric | Status | Date | |
|---|---|---|---|---|---|---|
| LiveCodeBench Pro | Coding | 1299 | Codeforces Elo | Official | Aug 5, 2025 | Details |
| SWE-bench Pro | Coding | 16.20% | % resolved | Official | Aug 5, 2025 | Details |
| AIME (OTIS Mock) | Reasoning | 88.9% | accuracy | Official | Aug 5, 2025 | Details |
| GPQA Diamond | Reasoning | 75.8% | accuracy | Official | Aug 5, 2025 | Details |
| SimpleQA Verified | Other | 13.9% | accuracy | Official | Aug 5, 2025 | Details |
| LMArena | Chat preference | 1365 | source-defined rating | Official | Aug 5, 2025 | Details |
| WeirdML | Coding | 48.2% | average accuracy | Official | Aug 5, 2025 | Details |
| Artificial Analysis Intelligence Index | Reasoning | 33.3 | Index | Unverified | Aug 5, 2025 | Details |
| Epoch Capabilities Index | Reasoning | 140.8 | Index | Official | Aug 5, 2025 | Details |
| Aider Polyglot | Coding | 41.8% | % correct | Official | Aug 5, 2025 | Details |
| MMLU-Pro | Reasoning | 80.8% | accuracy | Verified | Aug 5, 2025 | Details |
| τ²-bench (Telecom) | Tool use | 65.8% | pass^1 | Official | Aug 5, 2025 | Details |
| GDPval | Agents | 947 | Elo | Official | Aug 5, 2025 | Details |
| LiveCodeBench | Coding | 87.8% | Pass@1 | Unverified | Aug 5, 2025 | Details |
| SciCode | Coding | 38.9% | accuracy | Unverified | Aug 5, 2025 | Details |
| AA-Omniscience: Knowledge and Hallucination Benchmark | Reasoning | -50 | AA-Omniscience Index | Official | Aug 5, 2025 | Details |
| IFBench | Reasoning | 69.0% | accuracy | Official | Aug 5, 2025 | Details |
| MultiChallenge | Reasoning | 45.34% | accuracy | Verified | Aug 5, 2025 | Details |
| Global-MMLU | Reasoning | 82.8% | accuracy | Unverified | Aug 5, 2025 | Details |
| EQ-Bench Creative Writing v3 | Chat preference | 1041 | Elo | Verified | Aug 5, 2025 | Details |
| Design Arena | Chat preference | 1017 | Elo | Verified | Aug 5, 2025 | Details |
| MASK (Model Alignment between Statements and Knowledge) | Other | 92.00 | Honesty score | Verified | Aug 5, 2025 | Details |
| MCP-Universe | Tool use | 25.54% | Overall Success Rate | Verified | Aug 5, 2025 | Details |
| Vectara Hallucination Leaderboard | Other | 14.2% | Hallucination Rate | Official | Aug 5, 2025 | Details |
| MultiNRC | Reasoning | 15.17% | accuracy | Official | Aug 5, 2025 | Details |