API Reference

Every consumer-facing operation, request schema, and response shape in the Civitai Orchestration API. Pages here are generated from the OpenAPI specification (v2-consumers.json) and stay in sync with the running API on every build.

Conventions

• Base URL: https://orchestration.civitai.com
• Auth: Authorization: Bearer <token> on every request.
• Content type: application/json for bodies; blob upload endpoints accept multipart/form-data or presigned PUT.
• IDs: workflow IDs are ULIDs prefixed wf_; blob IDs are prefixed blob_.
• Polymorphism: workflow step bodies use a $type discriminator; request/response schemas list all valid subtypes under oneOf.

Entry points

Most consumer integrations only touch three operations:

• SubmitWorkflow — create a workflow with one or more steps
• GetWorkflow — poll a single workflow
• QueryWorkflows — list / filter workflows

The left sidebar is grouped by OpenAPI tag — Workflows, WorkflowSteps, Recipes, Blobs, Resources. Recipes have per-endpoint variants (one per job type) if you prefer the typed surface over the polymorphic SubmitWorkflow body.

Rate limits & quotas

Responses report the limits that apply to you in the RateLimit-Policy and RateLimit headers, using the named-policy syntax of the IETF RateLimit header fields draft. Each limit is one named entry:

RateLimit-Policy: "buzz";q=1000;w=60
RateLimit: "buzz";r=940;t=12

Parameter	Header	Meaning
q	RateLimit-Policy	Quota per window
w	RateLimit-Policy	Window length in seconds
r	RateLimit	Remaining quota in the current window
t	RateLimit	Seconds until the current window resets

When a limit is exhausted the API responds with 429 Too Many Requests and a Retry-After header giving the seconds to wait.

Free tier

A limited set of models can be run without Buzz on a best-effort basis. GetFreeTier returns the models currently offered and your remaining quota for each. To opt in:

• set "tier": "free" on SubmitWorkflow
• add ?tier=free to a recipe endpoint
• send X-Civitai-Tier: free on the OpenAI-compatible chat completions endpoint

Free work is scheduled behind paid work and expires if it cannot be served in time. Each free request uses quota when it is accepted, whether or not it later succeeds. Every step in a free workflow must use a model offered on the free tier; otherwise the request is rejected with 400. On free requests, each quota also appears as its own entry in the RateLimit headers.