TraceBench
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
TraceBench Agentic Coding
90.7Overall Score
4
TraceBench Function Calling
91.3Overall Score
4
TraceBench Deep Research
91.3Overall Score
4
TraceBench Agentic Coding SWE-bench
89Overall Score
4
TraceBench Function Calling Tau2Bench
77Overall Score
4
TraceBench Deep Research BrowseComp
81Overall Score
4