Returns public-safe model metadata and capabilities. startable: true means POST /processes in V1 accepts the model; other models (avatars, motion control, and the Suno music model) appear in process and asset reads but cannot be started yet. Audio model capabilities include voice/reference modes and setting ranges.
Authorization
bearerAuth models:readIn: header
Scope: models:read
Response Body
application/json
application/json
application/json
application/json
curl -X GET "https://example.com/models"{ "data": [ { "id": "string", "type": "string", "displayName": "string", "outputType": "image", "enabled": true, "startable": true, "schemaUrl": "string", "capabilities": {}, "pricing": { "unit": "image", "creditsByResolution": { "property1": 0, "property2": 0 }, "creditsByLength": { "property1": 0, "property2": 0 }, "pricedResolution": 0, "note": "string", "creditsPerJob": 0 } } ]}{ "error": { "code": "string", "message": "string", "requestId": "string" }}{ "error": { "code": "string", "message": "string", "requestId": "string" }}{ "error": { "code": "string", "message": "string", "requestId": "string" }}List spend per metadata value GET
Credits per value of one metadata key given at POST /processes, net of refunds: with metadata.client on every start, key=client is the cost per client. Covers the processes the key's member can read (the whole workspace for owners and admins). Values are returned as text. Up to 500 values, highest spend first.
List processes GET
Lists team processes visible to the API key. Supports at most one of status, type, or model; sessionId combines with status only. createdBy (comma-separated member ids, or me) and the createdAfter/createdBefore window combine with any of them. Unsupported filter combinations return unsupported_filter_combination instead of silently adding new index requirements.