/

/OpenRelay API

APIs
Generators
Stacks
Projects

Docs
History
Schema
Sources

v2

latestOpenAPI 3.1.02026-08-07139158192.4 KB

Internal

Liveness probe (no auth)

get/health

Response

Service is up

servicestring required

statusstring required

All 139 operations

get Liveness probe (no auth)
get List serving deployments (platform admin)
post Create a self-hosted serving deployment (platform admin)
get Serving deployment detail (platform admin)
delete Stop a serving deployment (platform admin)
post Drain a serving deployment (platform admin)
post Set the deployment backend's health (the emergency kill switch)
post Set one pool member's ranking weight (platform admin)
post Set the deployment pool's member-selection routing (platform admin)
get Per-replica runtime view of a serving deployment (platform admin)
post Set a serving deployment's desired replica count (platform admin)
post Set the deployment backend's routing weight (platform admin)
get Pre-deployment checks for a serving spec (platform admin)
get List published weights refs (platform admin)
get List batches, newest first
post Create a batch inference job
get Get a batch by id
post Cancel a running batch
get Get a cluster by id
get Get a cluster with its replicas and GPU model
post Restart a cluster
post Scale a cluster's replica count
post Stop a cluster
post Terminate a cluster
post Upload a JSONL input file for the Batch API
post Mint a signed URL for a direct JSONL upload
get Get a file's metadata
post Register a presigned upload after the PUT succeeds
get Download a file's raw JSONL content
get GPU availability by model
get GPU model catalog
get Locations / regions
get Current user profile and org memberships
get Pending org invites addressed to the signed-in user's email
post Accept a pending org invite (session only)
post Decline a pending org invite (session only)
patch Update the caller's profile
get List models in the OpenRelay catalog (OpenAI-compatible)
post Self-enroll a node using a provisioning token (no API key)
get List a node's reservations (platform admin)
post Reserve GPU units on a node for a set of orgs (platform admin)
get Get one node reservation (platform admin)
patch Amend a reservation's term, audience, or topology pin (platform admin)
post End a reservation now (platform admin)
post Take released GPU units back into a reservation (platform admin)
post Release GPU units back to the market for public sale (platform admin)
post Create the caller's profile and first org (idempotent)
post Create an organization (caller becomes owner)
get Get an organization
delete Delete (soft-delete) an organization (owner only)
patch Update an organization (owner/admin)
get List an org's API keys (no secrets)
post Create an API key (returns the plaintext once)
delete Revoke an API key
get Read the org's auto-recharge policy
put Set the org's auto-recharge policy
get Org balance + recent transactions
post Charge the saved card to top up prepaid balance
get List saved cards (brand/last4/exp only)
delete Remove a saved card
post Set a saved card as the default for charges
post Start saving a card (Stripe SetupIntent for inline Elements)
get List an org's clusters (cursor-paginated)
post Create a cluster
get List an org's pending invites (owner/admin)
delete Revoke a pending invite (owner/admin)
get List an org's members
post Invite a member by email (owner/admin)
delete Remove a member (owner/admin)
patch Update a member's role (owner/admin)
post Submit a provider application
get Provider earnings from usage records (+ payout history)
get List the provider org's nodes (with GPUs + location)
patch Set a node's operator lifecycle status (drain / maintenance / remove / online)
post Terminate spot workloads running on a node your org owns
get Provider node + token stats
get Provider status (isProvider + application status)
get List the org's provisioning tokens (provider only)
post Generate a one-time provisioning token (owner/admin, provider only)
delete Revoke a provisioning token (owner/admin)
get List an org's registry credentials (no secrets)
post Create a registry credential (credentials stored encrypted)
delete Delete a registry credential
get What GPU capacity is reserved to this org, and until when
get List runner pools (with active/queued usage)
post Create a runner pool
get GitHub connect URL (OAuth authorize when configured, else App install)
post List GitHub installations the user can connect (OAuth code exchange)
get Get a runner pool (with usage)
delete Delete (or drain) a runner pool
get List runner jobs for a pool
get Runner pool job metrics (last 30 days) + cost comparison
post Resize a runner pool's RAM tier
get List an org's VM snapshots (optionally filtered by source VM)
get List an org's SSH public keys
post Add an SSH public key (computes fingerprint)
delete Delete an SSH public key
post Initiate a resource transfer to another org (owner/admin)
get Pending incoming transfers (enriched)
get Outgoing transfers, all statuses (enriched)
get Count of pending incoming transfers
get Current-month usage (cents/hours + per gpuModel-tier breakdown)
get Per-day spend over a trailing window (compute + inference segments)
get List an org's VMs (cursor-paginated)
post Create a VM
post Fork a new VM from a snapshot
get List an org's webhooks (no secrets)
post Create a webhook (returns the signing secret once)
delete Delete a webhook
patch Update a webhook (name/url/events)
get Recent delivery attempts for a webhook
post Regenerate a webhook's signing secret (returned once)
post Enable/disable a webhook
get List an org's pods (cursor-paginated)
post Create a pod
get Live pod GPU capacity and per-class rates
get GPU + CPU pricing
get One-command node onboarding installer (auth = org API key bearer, OR a pre-minted provisioning token via ?token=)
delete Soft-delete a snapshot
get Orgs the caller belongs to (for the transfer target dropdown)
post Accept a pending incoming transfer (target org owner/admin)
post Cancel a pending outgoing transfer (source org)
post Reject a pending incoming transfer (target org owner/admin)
get VM template catalog
get Get a VM by id
get Live burn rate, session cost so far, and org runway for a VM
get Get a VM with its GPU/node/template info and price
post Reboot a VM
post Resize a stopped VM's disk
post Restart a VM
post Create a snapshot of a VM
get SSH keys attached to a VM + the org's keys
post Attach an org SSH key to a VM
delete Detach an org SSH key from a VM
post Stop a VM
get Get a VM's embedded Telegram link
post Terminate a VM
post Set a VM's endpoint visibility
get The current principal and the org it acts in