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

# Search Models

> Search for available Ollama models with optional filtering and sorting

## Query Parameters

<ParamField query="query" type="string" default="">
  Search query string
</ParamField>

<ParamField query="categories" type="array">
  Filter by model categories

  <Expandable title="Available categories">
    * `embedding` - Embedding models
    * `cloud` - Cloud-optimized models
    * `vision` - Vision/multimodal models
    * `tools` - Tool-use models
    * `thinking` - Reasoning models
  </Expandable>
</ParamField>

<ParamField query="order" type="string" default="newest">
  Sort order for results

  * `newest` - Sort by newest first
  * `popular` - Sort by popularity (downloads)
</ParamField>

## Response

<ResponseField name="success" type="boolean">
  Indicates if the search was successful
</ResponseField>

<ResponseField name="models" type="array">
  Array of model objects matching the search criteria

  <Expandable title="model properties">
    <ResponseField name="name" type="string">
      Model name
    </ResponseField>

    <ResponseField name="description" type="string">
      Model description
    </ResponseField>

    <ResponseField name="tags" type="array">
      Array of model tags
    </ResponseField>

    <ResponseField name="downloads" type="number">
      Number of downloads
    </ResponseField>

    <ResponseField name="lastUpdated" type="string">
      Last updated timestamp
    </ResponseField>
  </Expandable>
</ResponseField>

<RequestExample>
  ```bash cURL theme={null}
  curl --request GET \
    --url 'http://localhost:3000/api/v1/ollama/search?query=llama&order=popular&categories=tools' \
    --header 'Authorization: Bearer <token>'
  ```

  ```javascript JavaScript theme={null}
  const response = await fetch(
    'http://localhost:3000/api/v1/ollama/search?query=llama&order=popular&categories=tools',
    {
      method: 'GET',
      headers: {
        'Authorization': 'Bearer <token>'
      }
    }
  );

  const data = await response.json();
  ```

  ```python Python theme={null}
  import requests

  url = "http://localhost:3000/api/v1/ollama/search"
  headers = {"Authorization": "Bearer <token>"}
  params = {
      "query": "llama",
      "order": "popular",
      "categories": ["tools"]
  }

  response = requests.get(url, headers=headers, params=params)
  data = response.json()
  ```
</RequestExample>

<ResponseExample>
  ```json 200 - Success theme={null}
  {
    "success": true,
    "models": [
      {
        "name": "llama2",
        "description": "Meta's Llama 2 model for general chat and text generation",
        "tags": ["chat", "7b", "13b", "70b"],
        "downloads": 5420000,
        "lastUpdated": "2024-01-10T15:30:00Z"
      },
      {
        "name": "llama2-uncensored",
        "description": "Uncensored version of Llama 2",
        "tags": ["chat", "uncensored"],
        "downloads": 890000,
        "lastUpdated": "2024-01-08T10:00:00Z"
      }
    ]
  }
  ```

  ```json 401 - Unauthorized theme={null}
  {
    "success": false,
    "error": "Unauthorized"
  }
  ```

  ```json 500 - Server Error theme={null}
  {
    "success": false,
    "error": "Search failed"
  }
  ```
</ResponseExample>

## Notes

* Leave `query` empty to browse all models
* Multiple categories can be specified to filter results
* Results are paginated automatically by the Ollama registry