CursorBench
CursorBench: Real-World IDE Code Editing and Navigation
Practical benchmark measuring an AI model's ability to perform code edits, fast diff generation, symbol resolution, and context retrieval in realistic editor workflows.
Verified Model Leaderboard
Ranked by verified Score (%) score from official technical reports.
Evaluation Methodology & Harness
Evaluates structured edit diff accuracy, adherence to repository style, and multi-file contextual updates across production codebases.
Hundreds of realistic code modification scenarios and refactoring tasks in active open-source projects.
Why This Benchmark Matters
Directly reflects day-to-day developer productivity in AI code editors like Cursor, VS Code, and Windsurf.
Known Limitations & Contamination Risks
Fast-moving benchmark dependent on editor client orchestration harness.
Data Accuracy & Verification Notice
AI model specifications, pricing records, and benchmark metrics published on this platform are compiled directly from authoritative sources (official provider documentation, research papers, and verified evaluation harnesses). Benchmark results reflect specific test harnesses and prompting methodologies; scores are not directly comparable across differing evaluation setups.