Open Code Review AACR-Bench - Precision: leaderboard
Precision on Open Code Review's AACR-Bench run: the share of a reviewer's posted comments that match an expert-verified issue at the right lines. Rewards quiet, accurate reviewers; a harness that comments on everything scores low here and high on recall.
Metric: Precision (%). Source: open-codereview.ai. Saturation forecast: Not forecast. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GLM-5.2 | 32.3 | #80 |
| 2 | GPT-5.5 | 32.1 | #33 |
| 3 | DeepSeek V4 Pro | 30.6 | #112 |
| 4 | GLM-5.1 | 28.9 | #109 |
| 5 | Claude Opus 4.8 | 15.93 | #39 |
| 6 | Qwen 3.7 Max | 8.23 | #85 |
| 7 | Claude Opus 4.6 | 7.23 | #67 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-code-review-aacr-bench-precision · How It Works · Data refreshed daily, snapshot 2026-11-05.