codecompanion-ai/swe-bench
Open sourced predictions, execution logs, trajectories, and results from model inference + evaluation runs on the SWE-bench task.
Open sourced predictions, execution logs, trajectories, and results from model inference + evaluation runs on the SWE-bench task.