CursorBench (Multi-file Editing) Leaderboard
Evaluates AI coding assistants on real-world multi-file editing tasks across diverse codebases. Tests the ability to understand project context and make coordinated changes.
Why it matters: The first benchmark designed specifically for agentic coding assistants that edit multiple files. More realistic than single-function benchmarks like HumanEval.
Top Model
61.3%
Composer 2
Average Score
61.3%
Across 1 model
Models Tested
1
Metric: pass rate
Human Baseline
-
Score Range: 0%–100%
CursorBench Scores - Top 1 Models
Ranked by CursorBench score (%)
Model Rankings
All models with a reported CursorBench score, ranked by highest pass rate.
About CursorBench
- Full Name
- CursorBench (Multi-file Editing)
- Category
- Coding
- Metric
- pass rate (%)
- Score Range
- 0%–100%
- Human Baseline
- Not established
- Status
- Active
CursorBench is a standardized evaluation that measures AI model performance on specific tasks. It provides comparable scores across different models, helping developers choose the right model for their needs.
Composer 2 currently holds the top score on the CursorBench benchmark. See our full rankings table above for the complete leaderboard with 1 models.
We update benchmark data from multiple sources including HuggingFace open-source model leaderboards and LMArena. Scores are refreshed regularly as new evaluations are published and new models are released.
No. While CursorBench is an important indicator, real-world performance depends on many factors including pricing, latency, context window, and specific task requirements. We recommend using our composite score which weighs multiple benchmarks and practical factors.