DeepSWE-Preview
State-of-the-art RL-trained coding agent for complex SWE tasks
... scaling, and innovative policy gradient strategies adapted from GRPO, DAPO, Dr.GRPO, and RLOO. DeepSWE-Preview showcases strong reasoning, file navigation, and patch submission skills. It is ideal for agent-based code repair, debugging, and PR generation across real-world repositories. The model is served using platforms like vLLM and Hugging Face TGI, with support for 64k context length and OpenAI-compatible APIs.