BridgeBench Debugging — leaderboard

AI debugging benchmark: 10 tasks measuring ability to diagnose and fix broken TypeScript code, testing reproduction, regression resistance, and diagnostic accuracy.

Metric: Confidence-weighted Elo. Source: www.bridgebench.ai. 2 models tracked.

Top models

#ModelScore
1Gemini 3.1 Pro (Preview)1004.1
2Kimi K2.7 Code995.9

Interactive version: theaggregate.ai/benchmark?slug=bridgebench-debugging · How the rankings work · Data refreshed daily, snapshot 2026-07-22.