Skip to main content

Model Management

NineData's model management capabilities allow administrators to configure, save, and maintain custom AI models on the platform. Custom models can be seamlessly integrated into all AI-driven features of NineData, such as TEXT2SQL conversion, object code generation, ChatDBA chatbot, or knowledge base vectorization, to enhance data processing efficiency and customization capabilities.

Background Information​

NineData uses its own database models to support the AI capabilities of the NineData platform by default. To meet users' needs for different models, NineData has introduced the model management feature. Through this feature, users can enter other large models into the NineData platform and allow all AI functions within the NineData platform to use these large models.

Prerequisites​

  • Logged in to the NineData platform with an account that has system administrator privileges.
  • An API Key has been created on the target model service provider's page.

Adding Custom Models​

In the absence of any added models, NineData will use its own models to drive all AI functions. You can replace the original models by adding custom models through this step.

  1. Log in to the NineData Console.

  2. In the left navigation bar, click .

  3. Click the in the upper right corner of the page.

  4. Configure the following parameters according to the table below.

    ParameterDescription
    The unique name that identifies the custom model, used for quick identification in the model list.
    Example: DeepSeek-R1
    Select the service source of the model, currently supports DeepSeek and Qwen.
    Specify the API call URL of the model, which can be obtained from the official page of the model supplier.
    Example: https://api.deepseek.com/v1/chat/completions.
    API KeyEnter the API Key of the model, used for identity authentication when calling the model API, which can be obtained from the official page of the model supplier.
    Specify the underlying AI engine to be used, affecting the model's capabilities and functional compatibility.

  5. Click .

Disabling Custom Models​

You can disable custom models at any time and revert to using NineData's built-in models.

  1. Log in to the NineData Console.

  2. In the left navigation bar, click .

  3. In the list, click the under the column on the right side of the target model, and then click in the pop-up confirmation window.

Enabling Custom Models​

You can re-enable a custom model after disabling it. You can also choose to enable any one of the custom models among multiple custom models.

  1. Log in to the NineData Console.

  2. In the left navigation bar, click .

  3. In the list, click the under the column on the right side of the target model, and then click in the pop-up confirmation window.

    tip
    • Only one model can be enabled at a time.
    • After disabling all custom models, NineData's built-in model will be used by default.

Editing Custom Models​

If there are changes in the information of a custom model, you can edit the parameters of that custom model.

  1. Log in to the NineData Console.
  2. In the left navigation bar, click .
  3. In the list, click the more> under the column on the right side of the target model.
  4. After completing the editing on the page, click .

Deleting Custom Models​

If a custom model is no longer needed, it can be manually deleted.

  1. Log in to the NineData Console.
  2. In the left navigation bar, click .
  3. In the list, click the more> under the column on the right side of the target model, and then click in the pop-up confirmation window.
Last updated on