GPT-5.2
Benchmark scores
| Benchmark | Category | Rank | Score | Captured |
|---|---|---|---|---|
| METR Time Horizon | agents | #3 | 5h52m | 2025-12-11 |
| Terminal-Bench 2.0 | agents | #7 | 65.8% | 2026-02-10 |
| AA Coding Index | code | #7 | 48.7% | 2026-05-02 |
| SWE-bench Verified | agents | #11 | 72.8% | 2026-02-19 |
| Chatbot Arena | chat | #16 | 1439 | 2026-05-01 |
| OpenRouter · Weekly Usage | usage | #26 | #60 | 2026-05-02 |