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

SWE-Skills-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Software Engineering Task SuccessSWE-Skills-Bench Hard
Task Success Rate35
20
Software EngineeringSWE-Skills-Bench
Baseline Accuracy98
4
Showing 2 of 2 rows