AI Model Configuration

OpsAnt has a built-in AI assistant that connects to OpenAI-compatible large model services. Complete the configuration in "System Settings - AI Configuration".

Configuration Steps

1. Enable AI assistant: Turn on the "Enable AI assistant" switch
2. Fill in the API Base URL: For example, https://api.deepseek.com/v1
3. Fill in the API Key: The key for the model platform
4. Test and fetch models: Click "Test and fetch models" to pull the list of available models from the platform
5. Select the default model: Choose one of the fetched models as the default model
6. Save

Supported Model Services

Any OpenAI-compatible /chat/completions interface can be connected, including:

• DeepSeek
• Qwen (Qwen / DashScope compatible mode)
• Ollama (local deployment)
• OpenAI
• Other compatible services

Model Selection

• Default model: The model used by the AI assistant by default, set on the configuration page
• Switch within a session: You can temporarily switch to other models in the upper-right corner of the AI assistant dialog box

Using the AI Assistant

After configuration and enabling, use it through the "AI assistant" entry in the upper-right corner (or "AI assistant" in the left menu):

• Natural language conversation
• Session management (create, rename, delete)
• Deep thinking mode
• Page context awareness

FAQ

• Failed to fetch models: Please confirm that the API Key is correct and the Base URL is accessible
• Conversation error: Please confirm that the AI assistant is enabled and the API Key has been configured