Replicable-MARL/MARLlib
A multi-agent reinforcement learning library built on PyTorch and Ray/RLlib for training cooperative and competitive agent systems.
Not currently ranked — collecting fresh signals.
star history
MARLlib provides a unified repository for multi-agent reinforcement learning research and implementation. It builds on Ray and RLlib for distributed training and supports diverse multi-agent environments including cooperative, competitive, and mixed settings. The library implements various MARL algorithms and offers integrated visualization tools for training metrics.
Frequently asked
- What is Replicable-MARL/MARLlib?
- A multi-agent reinforcement learning library built on PyTorch and Ray/RLlib for training cooperative and competitive agent systems.
- Is MARLlib open source?
- Yes — Replicable-MARL/MARLlib is open source, released under the MIT license.
- What language is MARLlib written in?
- Replicable-MARL/MARLlib is primarily written in Python.
- How popular is MARLlib?
- Replicable-MARL/MARLlib has 1.3k stars on GitHub.
- Where can I find MARLlib?
- Replicable-MARL/MARLlib is on GitHub at https://github.com/Replicable-MARL/MARLlib.