Agentix-Project/Agentix

Research infra for creating RL environments, post-training, and evals.

Creating realistic RL environments requires collaboration between researchers, engineers, and domain experts across many dimensions: artifacts, environments tools, dynamism of the environment, reproducibility, and more. There is no open source framework for building these environments effectively. Until now.

Scalable RL for Any Agent and Sandbox.

Harness for running and evaluating AI agents against RL environments

Make envoriment to train policy using agentic rl.

Evaluate and improve models and agents using environments

Framework for evaluating and improving agents

Agentic Research and Evaluation Suite

The official Python SDK for Eval Protocol