The Farama Foundation is a 501c3 nonprofit organization designed to house major open source reinforcement
learning (“RL”) libraries in a neutral nonprofit body. We aim to provide standardization and long term
maintenance for these libraries, as well as improvements to their reproducibility,
performance, and quality
of life features.
We are a large international team in charge of many of the most installed libraries in the field, such as Gymnasium (formerly Gym) and PettingZoo, and are open to new projects when it's a good fit. Our team previously maintained OpenAI Gym for 18 months before development moved into Gymnasium. We are also working to develop new key pieces of software that are missing in the open source reinforcement learning ecosystem.
We want to improve the tools for academic research in reinforcement learning, to make it easier, faster,
more productive, and more scientifically rigorous.
We want to create open source RL tools that “just work”, reducing the engineering workload required to apply reinforcement learning to real world problems and making novel forms of automation feasible.
We want to make “RLops” a thing the way MLops is.
If you'd like to learn more about our goals, vision, history and projects in more detail, take a look at our blog.
If you'd like to contact us for official reasons, you can reach out to us at firstname.lastname@example.org, though if you'd just like to chat our discord server is the best way to reach us. We'd love to hear from you.