Multi-SWE-Bench (rust): leaderboard
Multi-SWE-Bench Rust slice evaluates agents on resolving Rust GitHub issues.
Metric: Score (%). Source: huggingface.co. 36 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | MopenHands + Claude-3.7-Sonnet | 16.1 | |
| 2 | MopenHands + Gemini-2.5-Pro | 14.64 | |
| 3 | MopenHands + Claude-3.5-Sonnet(Oct) | 12.13 | |
| 4 | MSWE-agent + Gemini-2.5-Pro | 10.04 | |
| 5 | MagentLess + OpenAI-o3-mini-high | 7.95 | |
| 6 | MagentLess + OpenAI-o1 | 7.11 | |
| 7 | MagentLess + Doubao-1.5-thinking | 7.11 | |
| 8 | MSWE-agent + Claude-3.5-Sonnet(Oct) | 6.69 | |
| 9 | MSWE-agent + Claude-3.7-Sonnet | 6.69 | |
| 10 | MagentLess + DeepSeek-R1 | 6.69 |
Interactive version: theaggregate.ai/benchmark?slug=multi-swe-bench-rust · How It Works · Data refreshed daily, snapshot 2026-10-20.