New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
SWE-bench Pro benchmark
AI model leaderboard for the SWE-bench Pro benchmark. Compare how large language models score on SWE-bench Pro, see the full ranking, and understand what this AI benchmark measures. Claude Fable 5 currently leads with 80%. Harder professional-grade SWE-bench variant with multi-file refactors and stricter test suites.
Leaderboard
Frequently asked questions about SWE-bench Pro
What is the SWE-bench Pro benchmark?
Harder professional-grade SWE-bench variant with multi-file refactors and stricter test suites.
How is the SWE-bench Pro benchmark scored?
SWE-bench Pro is scored using the resolved (%) metric, where a higher score is better. The maximum achievable score is 100.000. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.
Which AI model scores highest on SWE-bench Pro?
As of the latest reported scores on GenAIList, Claude Fable 5 achieves the highest result on SWE-bench Pro with a score of 80%.
Is a higher SWE-bench Pro score better?
Yes. On SWE-bench Pro a higher score indicates better performance, so models near the top of the leaderboard are the strongest.