Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/cloudwaddie/lmarenabridge/llms.txt

Use this file to discover all available pages before exploring further.

The /api/v1/models endpoint returns the models that LMArena Bridge has discovered from LMArena. The list is fetched at startup and automatically refreshed every 30 minutes. The response format mirrors the OpenAI models list so it works with any OpenAI-compatible client.

​
GET /api/v1/models

Returns all models with text, search, or image output capabilities. Stealth models (those without a public organisation) are excluded. Authentication: Optional. Pass Authorization: Bearer <api_key> if your bridge has API keys configured; otherwise the header may be omitted.

​
Example request

cURL
curl http://localhost:8000/api/v1/models \
  -H "Authorization: Bearer your_api_key"

​
Example response

200
{
  "object": "list",
  "data": [
    {
      "id": "gpt-4o",
      "object": "model",
      "created": 1716000000,
      "owned_by": "openai"
    },
    {
      "id": "claude-3-5-sonnet-20241022",
      "object": "model",
      "created": 1716000000,
      "owned_by": "anthropic"
    },
    {
      "id": "gemini-1.5-pro",
      "object": "model",
      "created": 1716000000,
      "owned_by": "google"
    },
    {
      "id": "llama-3.1-405b-instruct",
      "object": "model",
      "created": 1716000000,
      "owned_by": "meta"
    }
  ]
}

​
Response fields

​
object
string
Always "list".
​
data
object[]
Array of model objects.

​
Ollama-compatible endpoints

LMArena Bridge also exposes a small set of Ollama-compatible stubs so that clients like Open WebUI can probe the server and discover models without additional configuration.

​
GET /api/v1/api/tags

Returns the same model list in Ollama format.
cURL
curl http://localhost:8000/api/v1/api/tags \
  -H "Authorization: Bearer your_api_key"
200
{
  "models": [
    {
      "name": "gpt-4o",
      "model": "gpt-4o",
      "modified_at": "2024-01-01T00:00:00Z",
      "size": 0
    }
  ]
}

​
GET /api/v1/api/ps

Returns an empty list of running models (stub endpoint for client compatibility).
200
{
  "models": []
}

​
GET /api/v1/api/version

Returns the bridge version string.
200
{
  "version": "0.1.0"
}

​
Model refresh

The bridge fetches the model list from LMArena at startup and re-fetches it every 30 minutes via a background task (periodic_refresh_task). If you add a newly released model on LMArena and it does not appear immediately, wait for the next refresh cycle or restart the bridge.
The created timestamp in each model object reflects the time the list was generated by the bridge, not when the underlying model was released by its provider.

Build docs developers (and LLMs) love

Get started for freeTalk to us