Develop and run production workloads at scale as simply as running on your laptop.
Learn how to get started with RLlib for scalable reinforcement learning, from training to serving.Watch the video
Missed Ray Summit? All the sessions and tutorials are now available on-demand.Watch now
New to Ray? Jump-start your learning with this free hands-on tutorial on Ray Core.Watch the tutorial
No matter where you are in your Ray journey, you can boost productivity, speed time to market, and simplify your operations by running your distributed machine learning and Python workloads on Anyscale.
Anyscale augments your Ray development experience, making it effortless to scale from laptop to a cluster. Here are just a few capabilities to enable the infinite laptop experience:
Anyscale abstracts away servers, dynamically scaling up and down to meet demands
Pods crash. Disks die. Nodes reboot. Anyscale handles all failures gracefully, without impacting execution
Seamlessly move apps to production; the code and library bundling is automatically handled for you
Centrally monitor the health of your resources and workloads with out of the box Grafana dashboards
Tap into Ray’s rich machine learning ecosystem to easily develop, scale, and deploy sophisticated machine learning applications. Simplify your operations with Ray as a unified distributed framework for your end-to-end machine learning workflows.
Supercharge your training with Ray Tune, serving with Ray Serve, Reinforcement learning workloads with RLlib, or use one of the many 3rd party library integrations.
Sign up for our early access program to try out the infinite laptop experience on Anyscale.