Which tools do Claude, Codex and Cursor choose? We measured 17k runs to find out
We watched almost 17k sessions across different types of personas, with 1,163 prompt variations, 75 repositories and 3 coding agents (Claude Code, Codex, Cursor) actually implementing the solutions instead of just recommending one.
Summary
A detailed analysis of 17,000 coding sessions reveals how three AI coding agents—Claude Code, Codex, and Cursor—perform in real scenarios. Discover which tool stands out in implementing solutions based on diverse prompts and repositories.
Comments (148 of ) ⌛