Description

RAX (Robot Agent eXploration) is an open-source framework where a frozen LLM coding agent learns real-world robot tasks by exploring and writing its own control programs. It operates without human demonstrations or policy training, instead identifying successful decision sequences and compiling them into a modular skill library that can be reused across different robot embodiments.

Descriptions, tags, and model credits may be AI-generated or inferred from public sources and can be incomplete or wrong. Learn more.