Skip to main content

Documentation Index

Fetch the complete documentation index at: https://ray-preview.mintlify.app/llms.txt

Use this file to discover all available pages before exploring further.

Ray collects anonymous usage statistics by default. The data helps the project understand which features are used and which versions are running in the wild. No personal data, no source code, no cluster contents are collected.

​
What’s collected

  • Ray version and the libraries you use (Core, Data, Train, Tune, Serve, RLlib).
  • A randomly generated cluster ID.
  • High-level cluster shape (number of nodes, presence of GPUs).
  • Whether the cluster is running on Kubernetes or a known cloud provider.

​
What’s not collected

  • Names of your tasks, actors, files, models, datasets, or paths.
  • Logs, metrics, or traces.
  • Code, configuration, or environment variables.

​
Opt out

Set the environment variable on every node before ray start: 
export RAY_USAGE_STATS_ENABLED=0
Or pass --disable-usage-stats to ray start: 
ray start --head --disable-usage-stats

​
Verify

ray usage-stats status

​
Next steps

FAQ

Common cluster-management questions.

Configure

Cluster start-time configuration.