GPT-5.3 Codex (xHigh): benchmark results
GPT-5.3 Codex evaluated at the xhigh reasoning-effort setting. Provider: OpenAI. Released 2026-02-05. Access: API.
Aggregate Rating 1797 ± 15, rank #145 of 2168 rated models, from 81 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EVMbench | 51 | Unweighted mean of Detect/Patch/Exploit task means (%); no h | 100 |
| SWE-AGI (Hard) | 5 | Tasks passed (out of 8): the hard-tier SWE-AGI tasks whose f | 100 |
| SWE-AGI (Medium) | 8 | Tasks passed (out of 8): the medium-tier SWE-AGI tasks whose | 100 |
| SWE-AGI (Medium) - Test-Suite Pass Rate | 100 | Test-suite pass rate (%, 0-100): mean over the 8 medium-tier | 100 |
| AA-Omniscience Index - Software Engineering (SWE) - Kotlin | 84 | Omniscience Index | 98.1 |
| ALE-Bench | 1655.22 | Performance (Self-Refine x1) (imported) | 97.8 |
| AA-Omniscience Index - Software Engineering (SWE) - JavaScript | 82.73 | Omniscience Index | 97.6 |
| AA Long Context Reasoning | 83.33 | Accuracy (%) | 96.4 |
| AA Terminal-Bench Hard | 53.03 | Accuracy (%) | 96.2 |
| AA-Omniscience Index - Software Engineering (SWE) - Python | 80.5 | Omniscience Index | 96.2 |
| AA-Omniscience Index - Software Engineering (SWE) - TypeScript | 86.67 | Omniscience Index | 96.2 |
| AA-Omniscience Index - Software Engineering (SWE) - Julia | 76 | Omniscience Index | 96 |
Interactive version: theaggregate.ai/model?slug=gpt-5-3-codex-xhigh · How It Works · Data refreshed daily, snapshot 2026-10-20.