Codex Mini: benchmark results
Provider: OpenAI. Access: API.
Aggregate Rating 1659 ± 35, rank #565 of 10916 rated models, from 8 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BenchTable - Tech | 79.6 | Weighted Score (%) | 85 |
| SnakeBench | 28.1 | TrueSkill Rating | 80.4 |
| AI Chess Leaderboard (Reasoning) | 1108 | Elo | 76.6 |
| AI Chess Leaderboard (Continuation) | 835 | Elo | 70.4 |
| BenchTable - Reasoning | 56.5 | Weighted Score (%) | 70.1 |
| BenchTable | 58 | Total Score (%) | 68 |
| BenchTable - STEM | 59.5 | Weighted Score (%) | 62.1 |
| BenchTable - Utility | 55.3 | Weighted Score (%) | 54.9 |
Interactive version: theaggregate.ai/model?slug=codex-mini · How It Works · Data refreshed daily, snapshot 2026-11-08.