luchris429/purejaxrl
PureJaxRL is a high-performance end-to-end reinforcement learning implementation library built entirely in JAX.
Not currently ranked — collecting fresh signals.
star history
The library implements the entire RL training pipeline in JAX, including environments, enabling significant speedups through JIT compilation and parallel execution on GPUs. It supports running many agents in parallel for rapid hyperparameter tuning and meta-evolution for discovering new RL algorithms. The implementation includes PPO and other standard RL algorithms optimized for high-throughput training.
Frequently asked
- What is luchris429/purejaxrl?
- PureJaxRL is a high-performance end-to-end reinforcement learning implementation library built entirely in JAX.
- Is purejaxrl open source?
- Yes — luchris429/purejaxrl is open source, released under the Apache-2.0 license.
- What language is purejaxrl written in?
- luchris429/purejaxrl is primarily written in Python.
- How popular is purejaxrl?
- luchris429/purejaxrl has 1.1k stars on GitHub.
- Where can I find purejaxrl?
- luchris429/purejaxrl is on GitHub at https://github.com/luchris429/purejaxrl.