Contextbench/Tracebench download history

Contextbench/Tracebench is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 1,065 times (185 in the last 7 days), and 19,840 times in total. It ranks #18,438 among datasets by monthly downloads.

Tracebench This dataset contains agent trajectories (TerminalBench + SWE-bench) with two splits: full: 3316 trajectories (2670 terminal + 646 SWE-bench) verified: 1000 trajectories (489 SWE-bench + 511 terminal; terminal selected by step_count>=20, has incorrect steps, error-stage ratio

Open Contextbench/Tracebench on Hugging Face

Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.