OpenRL-Lab/openrl
A unified reinforcement learning framework built on PyTorch for training single and multi-agent systems.
Not currently ranked — collecting fresh signals.
star history
OpenRL is a unified reinforcement learning framework that supports training agents across diverse environments including games (Atari, Google Research Football), robotics (Mujoco), and embodied AI tasks. It provides distributed training capabilities, multi-agent reinforcement learning support, and integration with LLM-based agents and MCTS algorithms. The framework builds on PyTorch and supports Gym/Gymnasium environments.
Frequently asked
- What is OpenRL-Lab/openrl?
- A unified reinforcement learning framework built on PyTorch for training single and multi-agent systems.
- Is openrl open source?
- Yes — OpenRL-Lab/openrl is open source, released under the Apache-2.0 license.
- What language is openrl written in?
- OpenRL-Lab/openrl is primarily written in Python.
- How popular is openrl?
- OpenRL-Lab/openrl has 839 stars on GitHub.
- Where can I find openrl?
- OpenRL-Lab/openrl is on GitHub at https://github.com/OpenRL-Lab/openrl.