CHI-Bench (Codex): leaderboard
Metric: Pass@1 (%). Source: actava.ai. Saturation forecast: Around 2028. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.6 Sol | 25.33 |
| 2 | GPT-5.5 | 20.89 |
| 3 | GPT-5.4 | 16 |
| 4 | GPT-5.6 Luna | 13.33 |
| 5 | GPT-5.6 Terra | 13.33 |
| 6 | GPT-5.4 Mini | 8.44 |
Interactive version: theaggregate.ai/benchmark?slug=chi-bench-codex · How It Works · Data refreshed daily, snapshot 2026-10-09.