Model Configuration

API Address Configuration

When configuring a model provider, you need to correctly fill in the API address (Base URL).

Base URL Format Guidelines

Different providers have different API address formats:

Standard Format

Most providers follow this format:

https://api.example.com/v1

Common Provider Base URLs

• 302.AI: https://api.302.ai/v1
• OpenAI: https://api.openai.com/v1
• Anthropic: https://api.anthropic.com/v1
• Google Gemini: https://generativelanguage.googleapis.com/v1

Custom API Address

If you're using a self-hosted service or proxy service, you can fill in a custom API address:

1. Find the "Base URL" field in the provider configuration page
2. Enter your custom API address
3. Ensure the address is compatible with the provider's API format

Fetching Model List

After configuring the provider, you need to fetch the available model list.

Auto-Fetch Models

This is the recommended method, where the system automatically fetches the latest model list from the provider's API:

1. In the provider configuration page, ensure Base URL and API Key are correctly filled in
2. Click the "Fetch Models" button
3. Wait for the system to fetch the model list from the provider
4. After successful fetching, you'll see all available models

Auto-Update Model List

You can enable auto-update to have the system periodically fetch the latest models:

1. In the provider configuration page, find the "Auto-update model list on window open" option
2. Turn on this switch
3. The system will automatically update the provider's model list each time the application window is opened

Adding Models

Manually Add Models

If the provider doesn't list certain models, or if auto-fetch fails, you can manually add them:

1. In the provider configuration page, click the "Add Model" button
2. Fill in the following information:
   • Model ID: The unique identifier of the model (e.g., gpt-4-turbo)
   • Model Name: The name displayed in the interface (e.g., GPT-4 Turbo)
   • Model Description: Optional description
   • Model Type: Select the model type (chat/embedding/image, etc.)
3. Configure model capabilities (see below)
4. Click the "Save" button

Editing Models

For already added models, you can edit their configuration at any time.

Edit Model Information

1. Find the model you want to edit in the model list
2. Click the "Edit" icon on the right side of the model
3. You can modify the following:
   • Model ID: The unique identifier of the model
   • Model Name: Display name
   • Model Description: Description information
   • Model Type: Model type (chat/embedding, etc.)
   • Model Capabilities: Supported features
   • Enable Status: Whether available in conversations
4. Click "Save" after modifications

Model Capability Configuration

Each model has different capabilities. Correct configuration ensures proper functionality:

Vision Capability

If the model supports image input and understanding:

• Enable "Vision" capability
• Users can upload images in conversations
• The model can analyze and understand image content

Function Calling

If the model supports calling external functions and tools:

• Enable "Function Calling" capability
• The model can use predefined functions
• Supports more complex interaction scenarios

Reasoning Capability

If the model has enhanced reasoning capabilities:

• Enable "Reasoning" capability
• The model performs better on complex problems
• Suitable for tasks requiring logical analysis

Other Capabilities

• Music: Model supports music generation
• Video: Model supports video processing

Managing Models

Delete Models

If you no longer need a model:

1. Find the model you want to delete in the model list
2. Click the "Delete" icon on the right side of the model
3. Confirm the deletion

Common Issues