Sumanth Hegde of Anyscale Compute is developing native vLLM fixes for weight synchronization and asynchronous RL scaling

Excited to share some of our work on improving vLLM for RL! A number of RL frameworks, including SkyRL, use vLLM for inference, and we’ve noticed some common problems: 1. Weight syncing between training and inference is implemented in an ad-hoc fashion and duplicated across frameworks. 2. Asynchronous RL is prone to break at scale, especially in P/D and DPEP deployments. We’ve been working on improving both!