Quick BI lets you flexibly configure large models. You can select and switch between different large models for intelligent features, such as Q-Assistant, Q-Insight, and Q-Builder, to support various scenarios for intelligent interaction and data analytics. This topic describes how to configure large models.
Feature entry point
You can access the Large model configuration page as shown in the following figure.
Built-in large models
Quick BI supports the following built-in large models: DeepSeek-R1, QWEN-Max, and QWEN-Long.
Quick BI provides these built-in large models. To ensure data security, Quick BI does not record logs or store data.
Custom large models
Add a custom large model
If you want to use a third-party large model that you purchased, you can add it as shown in the following figure.
Click Add.
In the Add Third-party Large Model Service dialog box, configure the following parameters.
Parameter
Description
Service provider
Alibaba Cloud Model Studio is supported.
Model name
Select a built-in large model or enter a custom model name.
Context limit
The context length limit for the model. Valid values are 8k, 16k, 32k, 64k, and 1000k.
Display name
The display name for the custom large model. The name can be up to 50 characters long.
API key
The third-party large model's API key.
Click Test Connection to test the connection to the large model.
After the test succeeds, click Confirm to save the configuration.
After the model is added, you can set it as the default model or select it for use in features such as Q-Assistant and Q-Insight.
Manage custom large models
You can enable (①), disable (②), edit (③), or delete (④) a custom large model.
① If a model is disabled, click the 
icon to enable it.
② If a model is enabled, click the 
icon to disable it.
③ Click the 
icon to edit the model. You can only modify the Display name.
④ Click the 
icon to delete the model.
Default model
You can set a large model as the default model to avoid repetitive selections and improve efficiency.
Effective scope:
Applicable scenarios
After you set a default model, it is used by default in scenarios that allow model switching, such as Q-Insight.
For example, you can set QWEN-Max as the default large model.
It is then selected by default when you use Q-Insight. You can switch to other large models as needed. 
Non-applicable scenarios
To ensure the best analysis results, the Q-Assistant feature uses the built-in Official Inference model by default. Therefore, the default model setting does not apply to this feature. However, you can still switch to other large models as needed.
