AIThe New Stackabout 2 hours ago
GitHub Unveils ReviewBench AI Benchmark
6 min read
On Monday, GitHub unveiled ReviewBench, an open benchmark developed with Microsoft to measure AI code review agents. GitHub Copilot code review leads the inaugural leaderboard with a 40.1% grounded F1 score. The dataset uses 219 public pull requests from 187 repositories spanning 19 languages. While Martian's independent benchmark ranks Cubic first, GitHub states its tool consistently anticipated production improvements.
Level
Hype check
Tap to vote and see what everyone thinks.
