GET /v1/models

Returns every model currently served by an online provider belonging to you. OpenAI-format response — drop-in compatible with anything that calls /models to populate a dropdown.

Request

curl https://api.inference.club/v1/models \
  -H "Authorization: Bearer $INFERENCE_CLUB_KEY"

Response

{
  "object": "list",
  "data": [
    {
      "id": "qwen3-8b",
      "object": "model",
      "created": 1729960000,
      "owned_by": "home-rig"
    },
    {
      "id": "llama-3.1-8b-instruct",
      "object": "model",
      "created": 1729960500,
      "owned_by": "office-3090"
    }
  ]
}

owned_by is the provider name that's serving the model — useful when you have multiple providers. If two providers serve the same id, only one entry appears (first writer wins).

Empty list?

If data is empty, no agent on your account is currently advertising any model. Either:

• Your agent isn't running — start it.
• Your agent isn't reaching inference.club — check its INFERENCE_CLUB_URL env var and look at agent logs.
• More than 60 seconds have passed since the last heartbeat — the provider is marked offline.

The /providers/my-nodes page in the dashboard is the easiest way to see the heartbeat status of each of your agents.