SWE-Bench ProMax - Go (Mini-SWE-Agent): leaderboard
Metric: Resolve rate (%, Pass@1 on the 23 Go instances; 170 repository-level refactoring instances from real commits (average 11.4 modified files), resolved when the patch passes every test of the manually reviewed suite; mini-swe-agent scaffold, step limit 300, cost limit 10 US dollars per instance). Source: arxiv.org. Saturation forecast: Around June 2028. 6 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro | 26.1 |
| 2 | GLM-5 | 21.7 |
| 3 | Kimi K2.5 | 17.4 |
| 4 | Claude Sonnet 4.6 | 13 |
| 5 | GPT-5.2 | 13 |
Interactive version: theaggregate.ai/benchmark?slug=swe-bench-promax-go-mini-swe-agent · How It Works · Data refreshed daily, snapshot 2026-09-29.