Auto-parallelism strategy advisor for LLM training and inference. Enumerates valid parallel strategies, estimates memory, models throughput, and recommends optimal configurations.

The RL Bridge for LLM-based Agent Applications. Made Simple & Flexible.

SkyRL: A Modular Full-stack RL Library for LLMs

🔥🔥🔥 ICLR 2025 Oral. Automating Agentic Workflow Generation.