Client Apps
CherryStudio Configuration Guide
All-in-one AI assistant desktop client
Introduction
CherryStudio is a powerful AI assistant desktop application that supports multiple mainstream AI models (Claude, Gemini, GPT, etc.), providing developers and users with a unified AI interaction interface.
- Official Website: https://www.cherry-ai.com/
- Download: https://www.cherry-ai.com/download
Features
- ✅ Multi-Model Support: Supports Claude, Gemini, GPT, and other mainstream AI models
- ✅ Unified Interface: Manage all AI services in one application
- ✅ Custom API: Connect to custom API providers
- ✅ Cross-Platform: Supports Windows, macOS, Linux
- ✅ Local-First: Data stored locally for privacy protection
- ✅ Rich Features: Conversation management, model switching, parameter adjustment, etc.
Installation Steps
Download and Install CherryStudio
- Visit the CherryStudio download page
- Choose the installer for your operating system:
- Windows: Download the
.exeinstaller - macOS: Download the
.dmgimage file - Linux: Download
.AppImageor.debpackage
- Windows: Download the
- Follow the system prompts to complete installation
macOS Users Note
If you see a message saying "Cannot open because it is from an unidentified developer," go to System Preferences → Security & Privacy and allow it to open.
Get 元渊智能 API Key
Before configuring CherryStudio, you need to get an API Key from the 元渊智能 platform:
- Visit the 元渊智能 Console
- Log in to your account
- Create the appropriate token group:
- Claude models: Select the C-prefixed group
- Gemini models: Select the gemini group
- Save the generated API Key
| Model | Provider Type | API Address | Token Group |
|---|---|---|---|
| Claude | Anthropic | https://api.ai.soraliststudio.com/v1/messages | C-prefixed group |
| Gemini | Gemini | https://api.ai.soraliststudio.com/v1beta/models | gemini group |
Configure Claude Models
Enter Settings Page
- Open the CherryStudio application
- Click "Settings" or "Preferences" in the lower left corner
- Select "Model Configuration" or "API Configuration"
Select Claude Model Type
Choose the Claude model you need from the model list:
- Claude Sonnet 4.5:
claude-sonnet-4-5-20250929- Latest version, suitable for daily coding and conversations (recommended) - Claude Opus 4.5:
claude-opus-4-5-20251101- Best performance, suitable for complex tasks - Claude 3.5 Haiku: Fast, suitable for simple tasks
Available models depend on your token group permissions on the 元渊智能 platform.
Configure Claude API
Fill in the following information in the Claude configuration interface:
- Provider Type: Select
Anthropic - API Address:
https://api.ai.soraliststudio.com/v1/messages - API Key: Paste your Claude API Key from the 元渊智能 Console
- Model Name: Enter the model name selected above (e.g.,
claude-sonnet-4-5-20250929)
Configure Gemini Models
Select Gemini Model Type
Choose the Gemini model you need from the model list:
- Gemini 3 Flash Preview:
gemini-3-flash-preview- Latest version, fast and performant (recommended) - Gemini 3 Pro Preview:
gemini-3-pro-preview- High performance for complex tasks - Gemini 2.0 Flash: Quick response for simple conversations
- Gemini and Claude require different API Keys (different token groups)
- Ensure you have created the corresponding token group on the 元渊智能 platform
Configure Gemini API
Fill in the following information in the Gemini configuration interface:
- Provider Type: Select
Gemini - API Address:
https://api.ai.soraliststudio.com/v1beta/models - API Key: Paste your Gemini API Key from the 元渊智能 Console
- Model Name: Enter the model name selected above (e.g.,
gemini-3-flash-preview)
Getting Started
Create a New Conversation
- Click the "New Chat" button
- Select a configured model from the model selector
- Start chatting with AI
Switch Models
You can switch models anytime during a conversation:
- Click the model selector at the top
- Select another configured model
- Continue the conversation
Adjust Parameters
CherryStudio typically supports adjusting these parameters:
- Temperature: Controls randomness of replies (0-1)
- Max Tokens: Controls reply length
- Top P: Controls sampling range
Parameter Suggestions
- Programming tasks: Temperature 0.2-0.5 (more accurate)
- Creative writing: Temperature 0.7-0.9 (more creative)
- Daily conversations: Temperature 0.5-0.7 (balanced)
Best Practices
1. Choose the Right Model
Match tasks to models:
- Code Writing: Claude Sonnet 4.5 (
claude-sonnet-4-5-20250929) - Quick Chat: Gemini 3 Flash Preview (
gemini-3-flash-preview) - Complex Reasoning: Claude Opus 4.5 (
claude-opus-4-5-20251101) - Multimodal Tasks: Gemini 3 Pro Preview (supports images)
2. Manage API Usage
- Regularly check your balance at the 元渊智能 Console
- Create different API Keys for different purposes
- Avoid exposing API Keys in public places
3. Optimize Conversation Experience
- Use clear prompts
- Set appropriate context length
- Make good use of conversation history management
Comparison with Other Clients
| Feature | CherryStudio | Alma | Hapi |
|---|---|---|---|
| Interface Type | Desktop App | Desktop App | Web/PWA |
| Multi-Model Support | ✅ | ✅ | ✅ |
| Code Editing | Partial | ✅ | ✅ |
| Terminal Integration | ❌ | ✅ | ✅ |
| Remote Access | ❌ | ❌ | ✅ |
| Learning Curve | Low | Medium | Medium |
Selection Guide:
- Pure conversation needs: CherryStudio (clean interface, easy to use)
- Programming development: Alma or Hapi (more powerful features)
- Remote control: Hapi (unique feature)
FAQ
API Key Invalid?
- API Key entered incorrectly
- Wrong token group selected (C-prefixed group Key cannot be used for Gemini)
- Insufficient balance
Solution: Check if the API Key was fully copied, confirm the correct token group was created in the 元渊智能 Console, and check account balance.
Model List Empty?
- Confirm Claude API address is
https://api.ai.soraliststudio.com/v1/messages - Confirm Gemini API address is
https://api.ai.soraliststudio.com/v1beta/models - Check network connection
- Re-obtain API Key and ensure token group permissions are correct
Slow Response?
- Check network connection quality
- Try using a faster model (e.g., Gemini Flash)
- Clean or reduce conversation history
How to Use Multiple Models Simultaneously?
- Configure different model providers separately
- Select the corresponding model when creating a new conversation
- Create multiple conversation windows, each using a different model
Key Takeaways
- ✅ Different models use different API Keys (different token groups)
- ✅ Claude API address:
https://api.ai.soraliststudio.com/v1/messages - ✅ Gemini API address:
https://api.ai.soraliststudio.com/v1beta/models - ✅ Choose the right model for your task
- ✅ Regularly check your balance at the 元渊智能 Console
- ✅ Keep your API Keys secure
How is this guide?