jiangxinke/Harness-RL
436 53PythonUpdated 2 months ago
Agentic RAG R1 Framework via Reinforcement Learning
agenticgrpoharnessrl
Our verdict: worth borrowing ideas from
Implements a novel RL‑based multi‑agent harness (CAPO) and training pipeline that could be extracted for custom harness logic
Filed under agent harnesses in our directory.