FrameworkAI and agent development
AgentRL
Trains multi-turn, multi-task agents with asynchronous reinforcement learning on Ray.
- Stars
- 313
- Forks
- 25
- Updated
- Updated Jul 10, 2026
MIT
Trains multi-turn, multi-task agents with asynchronous reinforcement learning on Ray.
AgentRL separates agent reinforcement-learning training from environment deployment. Rollout, actor, and reference workers run on Ray to generate trajectories, update the policy, and maintain a frozen KL baseline, while a controller and task workers manage multi-turn environments over HTTP or gRPC. It is suited to reproducing the paper’s experiments or adding tasks, but training requires Ray, suitable CPU/GPU resources, and model configuration.
Resource types
Use cases
Combine interchangeable models, data sources, and tools in LLM applications.
Trace, evaluate, monitor, and manage LLM, agent, and ML workflows.
Runtime
Protocols & integrations
Capabilities
Audience
Public GitHub facts last synced Jul 10, 2026.
Build and run Python agents with graph-based workflows
Build type-safe Python agents with tools, structured outputs, and provider choice