Models

`GET /v1/models`

Lists every model the gateway currently serves: the local pool’s live catalogue (chat and embedding models) plus, only if cloud upstreams are configured, the cloud pool’s pinned catalogue.

Accepts either a Bearer key or the console session cookie, the only /v1/* route that accepts both.

curl https://ai.rcfi.co.ke/v1/models \
  -H "Authorization: Bearer tak_your_key_here"

Response

{
  "object": "list",
  "data": [
    { "id": "ministral-3:3b", "object": "model", "owned_by": "taifa-ai" },
    { "id": "gemma2:2b", "object": "model", "owned_by": "taifa-ai" },
    { "id": "llama3.2:3b", "object": "model", "owned_by": "taifa-ai" },
    { "id": "nomic-embed-text", "object": "model", "owned_by": "taifa-ai" },
    { "id": "openrouter/free", "object": "model", "owned_by": "taifa-ai" }
  ]
}

The local entries reflect whatever Ollama actually has loaded right now; the openrouter/free entry only appears if the gateway has cloud upstreams configured at all. See Local vs. cloud routing.