Scalable, State of the Art Reinforcement Learning

RLlib is an open-source library for reinforcement learning (RL), offering support for production-level, highly distributed RL workloads while maintaining unified and simple APIs for a large variety of industry applications. Designed for quick iteration and a fast path to production, it includes 25+ algorithms that are all implemented to run at scale and in multi-agent mode.

RLlib is used by industry leaders in many different industries and use cases such as climate control, industrial control, manufacturing and logistics, finance, gaming, automobile, robotics, boat design, and many others.

If you either have your problem coded (in Python) as an RL environment or own lots of pre-recorded, historic behavioral data to learn from, you will be up and running in only a few days.
RLlib is already used in production by industry leaders in many different verticals, such as climate control, industrial control, manufacturing and logistics, finance, gaming, automobile, robotics, boat design, and many others.

Why RLlib?

Learn what makes RLlib stand out and why it’s the reinforcement learning library of choice for companies like Wildlife Studios, Riot Games, McKinsey / QuantumBlack and others.

Vibrant community

Reinforcement learning is hard. Easily find code examples and connect with other developers and experts.


Iterate quickly without needing to rewrite again to go to production or scale to a large cluster

State of the art

Choose from the latest and greatest in reinforcement learning algorithms to find the one best suited for your problem. Enjoy multi-agent support in all.

Supports external simulators

Optimize your policies using an industry- or problem-specific external simulator. Connect simulations to Rllib via its PolicyServer/Client architecture.

Seriously fast

Experience really fast policy evaluation, with lower overhead than most algorithms.

Tap into Ray ecosystem

Find the perfect set of hyperparamters using Ray Tune. Serve your trained model in a massively parallel way with Ray Serve.

RLlib in 60 seconds

It only takes a few steps to get your first RLlib workload up and running on your laptop.

RLlib does not automatically install a deep-learning framework, but supports TensorFlow (both 1.x with static-graph and 2.x with eager mode) as well as PyTorch. Depending on your needs, make sure to install either TensorFlow or PyTorch (or both, as shown below):

pip install "ray[rllib]" tensorflow torch

For installation on computers running Apple Silicon (such as M1), please follow instructions here. To be
able to run our Atari examples, you should also install:

pip install "gym[atari]" "gym[accept-rom-license]" atari_py.

This is all you need to start coding against RLlib.

Getting Started with RLlib

At a high level, RLlib provides you with an Algorithm class which holds a policy for environment interaction. Through the algorithm’s interface, you can train the policy compute actions, or store your algorithms. In multi-agent training, the algorithm manages the querying and optimization of multiple policies at once.

Reinforcement Learning