Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Code Debugging on DebugBench

66.6Average Accuracy

MOCHA

55.877658.661361.44564.2287May 14, 2026May 15, 2026May 16, 2026May 17, 2026May 18, 2026May 19, 2026
Updated 14d ago

Evaluation Results

MethodLinks
2026.05
66.6---
65.5866.0468.1762.52
2026.05
64.9965.3368.8260.82
2026.05
62.4663.2566.1657.97
2026.05
61.5---
2026.05
61.5---
2026.05
61.5---
2026.05
61.5---
61.3561.6561.5960.8
2026.05
60.7961.8863.9956.51
2026.05
56.2957.495754.38