AgentRL

Multitask, multiturn RL for LLM agents; Ray-based scaling, rollout/actor workers—for teams that want to train agents, not just run them.

trainingpython

Stars: 364
Adoption surface: complex
Autonomy: headless
Recovery: resumable
License: ✅ open-source
Category: Multi-agent and orchestration

Repository ↗ Example: Async GRPO trainer ↗

Related in Multi-agent and orchestration

MetaGPT
autogen
crewAI
OpenManus
ChatDev
openai-agents-python

Maintain AgentRL? Add the badge

Paste this into your README so readers can find the list and see how AgentRL compares:

[![Best of Agent Harnesses](https://img.shields.io/badge/%F0%9F%8F%86_Best_of-Agent_Harnesses-5ac4bf)](https://github.com/RyanAlberts/best-of-Agent-Harnesses)