SWE-bench for your codebase — mine your merged PRs into local, contamination-free coding-agent benchmarks. Adapters: claude-code, aider (Opus 4.7 / GPT-5.5 / Sonnet 4.6 / Gemini 3.1 Pro).
benchmark developer-tools ai-agents aider llm-eval coding-agents agent-evals swe-bench gemini-3-1-pro claude-opus-4-7 gpt-5-5
-
Updated
Apr 30, 2026 - Python