Cherry Studio Integration Guide
This guide will help you integrate GPT-Load proxy service with Cherry Studio AI client, supporting complete configuration for OpenAI, Gemini, Gemini OpenAI compatible, and Anthropic channel types.
Prerequisites
Ensure you have successfully deployed and started the GPT-Load service, running on http://localhost:3001 by default
General Setup Steps
Service Setup Location
Access Settings Page
Click "Settings" → "Model Services" in the bottom left corner of Cherry Studio
Add New Service
Click the "Add" button at the bottom of the page (do not use existing services from the list)
Get Models After Configuration
Click the "Manage" button to get the model list and select the desired models
Important Reminder
All channel types follow the same setup process: Select provider type → Configure API address and key → Get model list. Only the provider type and API address configuration differ.
1OpenAI Channel Configuration
Step 1: Create Service
Creation Steps
Enter Service Name
Set an easily recognizable name for your service
Select Provider Type
OpenAISelect OpenAI from the provider type dropdown
Click image to enlarge
Step 2: Configure Service
Configuration Parameters
Configure API Key
sk-123456Use the proxy key configured in your GPT-Load
Set API Address
http://localhost:3001/proxy/openaiWhere "openai" is the group name configured in your GPT-Load
Get Model List
After configuration, click the "Manage" button to get and select the required models
Click image to enlarge
2Gemini Channel Configuration
Step 1: Create Service
Creation Steps
Enter Service Name
Set an easily recognizable name for your Gemini service
Select Provider Type
GeminiSelect Gemini from the provider type dropdown
Click image to enlarge
Step 2: Configure Service
Configuration Parameters
Configure API Key
sk-123456Use the proxy key configured in your GPT-Load
Set API Address
http://localhost:3001/proxy/geminiWhere "gemini" is the group name configured in your GPT-Load
Get Model List
After configuration, click the "Manage" button to get and select the required models
Click image to enlarge
3Gemini OpenAI Compatible Format
Step 1: Create Service
Using Gemini channel's OpenAI compatible interface format
Creation Steps
Enter Service Name
Set a name for your Gemini OpenAI compatible service
Select Provider Type
OpenAIUse OpenAI type to support compatible interface
Click image to enlarge
Step 2: Configure Service
Configuration Parameters
Configure API Key
sk-123456Use the proxy key configured in your GPT-Load
Set API Address
http://localhost:3001/proxy/gemini/v1beta/openai/Note: The address must end with "/" to prevent Cherry Studio from automatically adding v1 path
Get Model List
After configuration, click the "Manage" button to get and select the required models
Click image to enlarge
Critical Configuration Note
The API address must end with "/"! This is a Cherry Studio rule requirement to ensure v1 path is not automatically added, guaranteeing the compatible interface works properly.
4Anthropic (Claude) Channel Configuration
Step 1: Create Service
Creation Steps
Enter Service Name
Set an easily recognizable name for your Anthropic service
Select Provider Type
AnthropicSelect Anthropic from the provider type dropdown
Click image to enlarge
Step 2: Configure Service
Configuration Parameters
Configure API Key
sk-123456Use the proxy key configured in your GPT-Load
Set API Address
http://localhost:3001/proxy/anthropicWhere "anthropic" is the group name configured in your GPT-Load
Get Model List
After configuration, click the "Manage" button to get and select the required models
Click image to enlarge
Important Notes
Configuration Considerations
• Please replace the GPT-Load access address in the examples with your actual service address
• The group names in the path (such as openai, gemini, anthropic) need to match your actual configuration in GPT-Load
• Use your actual proxy key configured in GPT-Load, do not use the example sk-123456
• After configuration, remember to select the newly added model in the chat interface for testing