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

# Create custom model

Create a new custom model or update an existing one (upsert by `model_name`). Custom models allow you to define organization-specific configurations with custom pricing, capabilities, and provider associations.

<Note>
  **Upsert behavior**: If a model with the same `model_name` already exists in your organization, it will be updated with the new values.
</Note>

## Request body

<ParamField body="model_name" type="string" required>
  Unique model name within your organization. This will be used to reference the model in API calls.

  <Accordion title="Example">
    ```json theme={"system"}
    "model_name": "my-custom-gpt-4"
    ```
  </Accordion>
</ParamField>

<ParamField body="base_model_name" type="string">
  Base model to inherit properties from. Can be a global model (e.g., `"gpt-4"`) or another custom model.

  <Accordion title="Example">
    ```json theme={"system"}
    "base_model_name": "gpt-4"
    ```
  </Accordion>
</ParamField>

<ParamField body="display_name" type="string">
  Human-readable display name for the model.

  <Accordion title="Example">
    ```json theme={"system"}
    "display_name": "My Custom GPT-4"
    ```
  </Accordion>
</ParamField>

<ParamField body="custom_provider_id" type="integer | string">
  ID or `provider_id` of the custom provider to associate with this model. See [Custom Providers](/api-endpoints/manage/models/custom-providers/list).

  <Accordion title="Example">
    ```json theme={"system"}
    "custom_provider_id": "my-azure-provider"
    ```
  </Accordion>
</ParamField>

<ParamField body="provider_id" type="integer | string">
  Alternative to `custom_provider_id`. Either one can be used.
</ParamField>

### Pricing

<ParamField body="input_cost" type="float">
  Cost per 1M input tokens in USD.

  <Accordion title="Example">
    ```json theme={"system"}
    "input_cost": 0.025
    ```
  </Accordion>
</ParamField>

<ParamField body="output_cost" type="float">
  Cost per 1M output tokens in USD.

  <Accordion title="Example">
    ```json theme={"system"}
    "output_cost": 0.075
    ```
  </Accordion>
</ParamField>

<ParamField body="cache_hit_input_cost" type="float">
  Cost per 1M cached input tokens in USD.
</ParamField>

<ParamField body="cache_creation_input_cost" type="float">
  Cost per 1M cache creation tokens in USD.
</ParamField>

### Capabilities

<ParamField body="max_context_window" type="integer">
  Maximum context window size for the model.

  <Accordion title="Example">
    ```json theme={"system"}
    "max_context_window": 8192
    ```
  </Accordion>
</ParamField>

<ParamField body="streaming_support" type="integer">
  Streaming support. `0` for no, `1` for yes.
</ParamField>

<ParamField body="function_call" type="integer">
  Function calling support. `0` for no, `1` for yes.
</ParamField>

<ParamField body="image_support" type="integer">
  Image/vision support. `0` for no, `1` for yes.
</ParamField>

<ParamField body="supported_params_override" type="object">
  Override UI parameter support for this model in the Playground. See the model detail endpoint for current values.

  <Accordion title="Example">
    ```json theme={"system"}
    {
      "supported_params_override": {
        "temperature": {
          "name": "temperature",
          "type": "number",
          "default": 0.7,
          "min": 0,
          "max": 1.5
        }
      }
    }
    ```
  </Accordion>
</ParamField>

## Response

Returns the created or updated model object.

<ResponseExample>
  ```json 201 Created theme={"system"}
  {
    "id": "my-custom-gpt-4",
    "model_name": "my-custom-gpt-4",
    "display_name": "My Custom GPT-4",
    "base_model_name": "gpt-4",
    "affiliation_category": "custom",
    "is_called_by_custom_name": false,
    "input_cost": 0.025,
    "output_cost": 0.075,
    "cache_hit_input_cost": 0.0,
    "cache_creation_input_cost": 0.0,
    "max_context_window": 8192,
    "streaming_support": 1,
    "function_call": 1,
    "image_support": 0,
    "source": "db",
    "provider": {
      "id": "my-custom-provider",
      "provider_id": "my-custom-provider",
      "provider_name": "My Custom Provider",
      "organization_id": 123
    }
  }
  ```

  ```json 400 Bad Request theme={"system"}
  {
    "error": "Base model not found"
  }
  ```

  ```json 401 Unauthorized theme={"system"}
  {
    "detail": "Authentication credentials were not provided."
  }
  ```
</ResponseExample>

<RequestExample>
  ```python Python theme={"system"}
  import requests

  url = "https://api.keywordsai.co/api/models/"
  headers = {
      "Authorization": "Bearer YOUR_API_KEY",
      "Content-Type": "application/json"
  }
  payload = {
      "model_name": "my-custom-gpt-4",
      "base_model_name": "gpt-4",
      "display_name": "My Custom GPT-4",
      "custom_provider_id": "my-azure-provider",
      "input_cost": 0.025,
      "output_cost": 0.075
  }

  response = requests.post(url, headers=headers, json=payload)
  print(response.json())
  ```

  ```typescript TypeScript theme={"system"}
  const url = 'https://api.keywordsai.co/api/models/';
  const headers = {
      'Authorization': 'Bearer YOUR_API_KEY',
      'Content-Type': 'application/json'
  };

  const payload = {
      model_name: 'my-custom-gpt-4',
      base_model_name: 'gpt-4',
      display_name: 'My Custom GPT-4',
      custom_provider_id: 'my-azure-provider',
      input_cost: 0.025,
      output_cost: 0.075
  };

  const response = await fetch(url, {
      method: 'POST',
      headers: headers,
      body: JSON.stringify(payload)
  });

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

  ```bash cURL theme={"system"}
  curl -X POST "https://api.keywordsai.co/api/models/" \
    -H "Authorization: Bearer YOUR_API_KEY" \
    -H "Content-Type: application/json" \
    -d '{
      "model_name": "my-custom-gpt-4",
      "base_model_name": "gpt-4",
      "display_name": "My Custom GPT-4",
      "custom_provider_id": "my-azure-provider",
      "input_cost": 0.025,
      "output_cost": 0.075
    }'
  ```
</RequestExample>