gpt-oss-120b & gpt-oss-20b Model Card

Benchmark Model Rank Results
agents-on-taubench-airlinegpt-oss-120b–Accuracy: 49.2
agents-on-taubench-retailgpt-oss-120b–Accuracy: 67.8
code-generation-on-aider-polyglotgpt-oss-120b (High)–Accuracy: 44.4
code-generation-on-swe-bench-verifiedgpt-oss-120b (High)–Accuracy: 62.4
multi-task-language-understanding-on-mmmlugpt-oss-120b–Accuracy: 81.3