New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Multi-SWE-bench benchmark
AI model leaderboard for the Multi-SWE-bench benchmark. Compare how large language models score on Multi-SWE-bench, see the full ranking, and understand what this AI benchmark measures. Kimi K2 Thinking currently leads with 41.9. Multilingual software-engineering benchmark — extension of SWE-bench across multiple languages.
Leaderboard
| # | Model | Organization | Score | Variant | Source |
|---|---|---|---|---|---|
| #1 | Kimi K2 Thinking | Moonshot | 41.9 | thinking-w-tools | official ↗ |
| #2 | Kimi K2 | Moonshot | 33.5 | k2-0905-w-tools-cited-k2-thinking | official ↗ |
Frequently asked questions about Multi-SWE-bench
What is the Multi-SWE-bench benchmark?
Multilingual software-engineering benchmark — extension of SWE-bench across multiple languages.
How is the Multi-SWE-bench benchmark scored?
Multi-SWE-bench is scored using the resolution_rate metric, where a higher score is better. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.
Which AI model scores highest on Multi-SWE-bench?
As of the latest reported scores on GenAIList, Kimi K2 Thinking achieves the highest result on Multi-SWE-bench with a score of 41.9.
Is a higher Multi-SWE-bench score better?
Yes. On Multi-SWE-bench a higher score indicates better performance, so models near the top of the leaderboard are the strongest.