DeepSWE is a 32B software engineering agent that achieves 59% on SWE-Bench-Verified with test-time scaling (42.2% Pass@1). It tops the SWE-Bench leaderboard for open-weight models.
Trained on top of Qwen3-32B to search, view, and navigate codebases, using RL to improve software engineering task completion.See the DeepSWE example for instructions on reproducing this with rLLM.Released: July 2025