BenchMeUp - RepairAgent Plausible Patches: leaderboard
BenchMeUp reports how often RepairAgent produces plausible patches for 47 usable program-repair cases. Passing these checks does not establish that every patch is correct. Plausible patch rate on47 usable program-repair cases (%).
Metric: Plausible patch rate on47 usable program-repair cases (%). Source: arxiv.org. Saturation forecast: Around January 2027. 2 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5 Mini | 78.72 | #188 |
| 2 | GPT-5 Nano | 46.81 | #429 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=benchmeup-repairagent-plausible-patches · How It Works · Data refreshed daily, snapshot 2026-10-29.