OverEager-Bench (Codex CLI): leaderboard
Metric: Overeager rate (%; share of 500 benign coding scenarios in which a deterministic trap predicate fires for an out-of-scope write or sensitive read, seed 42; Codex CLI 0.90.0). Source: arxiv.org. Saturation forecast: Estimated already saturated. 4 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.6 | 5.4 |
| 2 | MiniMax-M2.7 | 6.6 |
| 3 | GLM-4.6 | 13.5 |
| 4 | GPT-5.3 Codex | 13.8 |
Interactive version: theaggregate.ai/benchmark?slug=overeager-bench-codex-cli · How It Works · Data refreshed daily, snapshot 2026-09-26.