AI Assistant

Instructions for use

AI technology is deeply integrated into various workflows, becoming a new engine of efficiency and intelligence, widely applied in areas such as smart office, development, services, and creative assistance. Geographic Information System (GIS) software typically has complex functions and high learning costs, urgently requiring AI to revolutionize interaction methods and improve efficiency.

The AI Assistant feature launched in SuperMap iDesktopX, based on advanced large language models, aims to lower the threshold of GIS usage, enhance user operation efficiency, and usher in a new era of intelligent interaction through natural language operations, automated processes, and intelligent guidance.

The AI Assistant provides intelligent support for different task types, the current version offers three core features:

  • Intelligent Q&A: Focuses on Q&A and execution of operation commands;
  • Intelligent Mapping: Creates and modify thematic map styles through natural language, achieving intelligent expression from layers to visualization;
  • Intelligent Modeling: Calls modeling tools based on user intent to intelligently build processing and analysis models.

Feature Condition Descriptions

  • Data Requirements: Requires internet access to AI model services.
  • Permission Requirements: Using third-party models requires providing a valid API key.
  • Environment Configuration: To use the AI Assistant feature, you need to download and configure the JDK 17 environment. For details, please refer to: AI Assistant Environment Configuration.

Supported starting from SuperMap iDesktopX 2025 version.

Feature Entry

  • View Tab ->View Group ->Check AI Assistant (After checking, the AI Assistant panel will be displayed on the right side of the interface)

Operation Steps

1. In the AI Assistant panel, select the task type. Currently, Intelligent Q&A, Intelligent Mapping, and Intelligent Modeling task types are provided. Choose the corresponding task type based on your needs.  

Task Type Capability Description Example Questions
Intelligent Q&A
  • Performs Q&A based on native large models
  • Supports executing common operations: including open workspace, open map, Import SHP File
  • Open ../specific path/world.smwu workspace
  • Open world map
  • Import ../specific path/Point.shp file
Intelligent Mapping
  • Requires selecting the layer to create thematic maps in the layer manager
  • Intelligently creates thematic maps, including: unique value thematic map, label thematic map, graudated color thematic map, statistical thematic map, graduated symbol thematic map, dot density thematic map, etc.
  • Supports understanding semantics, modifying fields and basic styles through commands, but currently does not support customizing complete color ramps.
  • Create a graudated color thematic map for the current earthquake points layer
  • Create a statistical thematic map for the current regional economy layer
  • Create a label thematic map for the current text layer
Intelligent Modeling
  • Recommends modeling tools and builds models based on business intent
  • How to quickly build GPA?
  • How to generate a farmland slope classification map using GPA?
  • How to repair minor overlaps between earth polygons?

 2. Initiate Session:  Click the " Initiate Session" button to pop up a new session window.  Or directly double-click the task type to enter the corresponding  session interface.      Based on the selected task type, the session interface displays example questions for that task type.     

3. Model Switching: Click the model name drop-down menu to switch the currently used model.

  • Displays the current default model. If the current model is not configured with an API key, the assistant will prompt an error.
  • Supported models include:     DeepSeek-V3, qwen-max (tong yi qian wen), GPT-4o-mini (OpenAI), hunyuan-turbos-latest (tencent hun yuan), Claude 3.5 Haiku (Anthropic), Gemini 2.5 Flash (Google);

4. ClickModel Settings button to enter the configuration interface, which supports adding, modifying, and deleting models.

  • Model Service Name: Currently includes seven built-in services: DeepSeek, tong yi qian wen, OpenAI, tencent hun yuan, Anthropic, Google (Gemini), Ollama. You can select from the built-in service names.
  • Model Name: Specific model name (e.g., GPT-4o-mini).
  • Server Address: Service URL address for the model API.
  • Key: Access key that users need to obtain from the model service platform.

5. ClickTask Switch button, to switch between Intelligent Q&A, Intelligent Mapping, and Intelligent Modeling task types. Switching task types will create a new session.

6. Enter Question and Send:

  • Type your question in the input box based on the selected task type.

  • Click Send button to initiate AI response.

  • If you wish to interrupt generation during sending, click Stop Generation button.

7. Click the Home button in the upper right corner of the panel to return to the AI Assistant homepage.

Notes:
  • Please conduct Q&A based on the selected task type; otherwise, the AI Assistant may not provide effective responses.

  • Model server address and key must be correctly configured; otherwise, effective Q&A cannot be performed.

Related Topics

AI Assistant Environment Configuration