wanlilll/weavebench-traj download history

wanlilll/weavebench-traj is an other dataset on the Hugging Face Hub. In the last 30 days it was downloaded 109 times (40 in the last 7 days), and 529 times in total. It ranks #107,003 among datasets by monthly downloads.

WeaveBench Trajectories Full agent rollouts reproducing WeaveBench (arXiv:2606.09426) Table C8 gpt-5.4 row. agent = gpt-5.4 · judge = gpt-5.5 (openclaw / agent-as-judge) · τ = 0.80 · num_envs = 8 · 114 tasks across 8 domains. 🚧 More trajectories will be pushed continuously (additiona

Open wanlilll/weavebench-traj on Hugging Face

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