List hosted OpenAI-compatible models.

Anonymous callers get the public network-wide catalogue: models with at least one live, route-eligible profile on a public machine. Callers that send a valid customer-owned hosted gateway key instead get only the models the machines on their own account serve right now — the key's node scope is respected, private machines are included, and the account's model and route-tier policy still filters the list. The key-scoped list is returned uncached. Node IDs, runtime IDs, endpoint URLs, prompts, outputs, customer IDs, tokens, and tunnel details are never exposed. No authentication is required for the public view.

GET /v1/api/models

Responses

200 OpenAI-compatible model list.

application/json

object string REQUIRED

Enum: list

data object[] REQUIRED

Array of:

id string REQUIRED

object string REQUIRED

Enum: model

created integer REQUIRED

Unix timestamp from the freshest observed benchmark, or 0.

owned_by string REQUIRED

Enum: b3iq

default OpenAI-compatible error with B3IQ metadata.

curl -X GET 'https://{controlPlaneHost}/v1/api/models'

const response = await fetch('https://{controlPlaneHost}/v1/api/models', {  method: 'GET'});const data = await response.json();console.log(data);

200 Response

{  "object": "list",  "data": [    {      "id": "<string>",      "object": "model",      "created": 123,      "owned_by": "b3iq"    }  ]}

default Response

{  "error": {    "message": "<string>",    "type": "<string>",    "param": "<string>",    "code": "<string>",    "b3iq": {      "status": 123,      "title": "<string>",      "detail": "<string>",      "code": "<string>",      "job_id": "<string>",      "billing_lookup": "<string>",      "idempotency_status": "<string>"    }  }}

GET /v1/api/models

Base URL

Response

Was this page helpful?

Previous OpenAPI

Next Run a hosted OpenAI-compatible chat completion.