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.

The State API exposes the same data the dashboard renders: lists of nodes, jobs, actors, tasks, and objects, with filters and rich detail.

​
Python client

from ray.util.state import list_actors, list_nodes, list_tasks, get_actor

list_nodes()
list_actors(filters=[("state", "=", "ALIVE")])
list_tasks(filters=[("state", "=", "FAILED")])
get_actor(id="abc...")

​
CLI

ray summary actors
ray list tasks --filter "state=FAILED"
ray list nodes
ray get actor abc...

​
REST

The dashboard exposes the same endpoints under /api/: 
curl http://<head>:8265/api/v0/actors

​
Filters

list_tasks(
    filters=[
        ("state", "=", "FAILED"),
        ("name", "match", "*train*"),
    ],
    limit=100,
)

​
Use cases

  • Custom dashboards built on top of Ray state
  • Alerting when an actor crashes too often
  • Debugging stuck pipelines (ray list tasks --filter "state=PENDING")
  • Capacity planning: aggregate per-job CPU/GPU usage over time

​
Next steps

Dashboard

Built-in UI on top of these APIs.

Metrics

Time-series data.