Multi-SWE-Bench (go): leaderboard
Multi-SWE-Bench Go slice evaluates agents on resolving Go-language GitHub issues.
Metric: Score (%). Source: huggingface.co. 36 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | MopenHands + Gemini-2.5-Pro | 12.62 | |
| 2 | MSWE-agent + Gemini-2.5-Pro | 9.81 | |
| 3 | MopenHands + Claude-3.7-Sonnet | 7.48 | |
| 4 | MopenHands + Claude-3.5-Sonnet(Oct) | 6.79 | |
| 5 | MagentLess + Gemini-2.5-Pro | 6.07 | |
| 6 | MopenHands + Doubao-1.5-thinking | 6.07 | |
| 7 | MSWE-agent + Claude-3.5-Sonnet(Oct) | 5.84 | |
| 8 | MagentLess + Claude-3.7-Sonnet | 5.84 | |
| 9 | MopenHands + Llama-4-Maverick | 5.61 | |
| 10 | MSWE-agent + Claude-3.7-Sonnet | 5.37 |
Interactive version: theaggregate.ai/benchmark?slug=multi-swe-bench-go · How It Works · Data refreshed daily, snapshot 2026-10-20.