> ## Documentation Index
> Fetch the complete documentation index at: https://docs.compose.market/llms.txt
> Use this file to discover all available pages before exploring further.

# Models

> List, retrieve, search, and price the generated model catalog.

Model routes use public catalog IDs. Duplicate upstream model names are resolved by the generated catalog, so callers send the public ID shown by these endpoints.

## Routes

| Method | Path                        | Auth | Description                                               |
| ------ | --------------------------- | ---- | --------------------------------------------------------- |
| `GET`  | `/v1/models`                | None | Native generated model catalog.                           |
| `GET`  | `/v1/models/all`            | None | Extended catalog when bundled.                            |
| `GET`  | `/v1/models/{model}`        | None | One native model card.                                    |
| `GET`  | `/v1/models/{model}/params` | None | Optional parameter metadata for one model.                |
| `POST` | `/v1/models/search`         | None | Search and filter native model cards.                     |
| `GET`  | `/api/models`               | None | Public catalog proxy backed by the models service.        |
| `GET`  | `/api/pricing`              | None | Compact pricing table from the bundled inference catalog. |

## Search body

<ParamField body="q" type="string">
  Search text matched against model metadata.
</ParamField>

<ParamField body="provider" type="string">
  Provider slug. Invalid providers return `400`.
</ParamField>

<ParamField body="modality" type="string">
  Canonical modality such as `text`, `image`, `audio`, `video`, or `embedding`.
</ParamField>

<ParamField body="operation" type="string">
  Canonical operation for the selected modality.
</ParamField>

<ParamField body="limit" type="number">
  Maximum models to return.
</ParamField>

<ParamField body="cursor" type="string">
  Cursor returned by the previous search page.
</ParamField>

## Response

<ResponseField name="object" type="string">
  `list` for list and search routes.
</ResponseField>

<ResponseField name="data" type="array">
  Model cards from the generated catalog.
</ResponseField>

<ResponseField name="next_cursor" type="string">
  Search cursor when more results are available.
</ResponseField>

<ResponseExample>
  ```json Response theme={null}
  {
    "object": "list",
    "data": [
      {
        "modelId": "gpt-5.5",
        "provider": "azure",
        "name": "gpt-5.5"
      }
    ],
    "total": 1,
    "next_cursor": null
  }
  ```
</ResponseExample>