CherryStudio Configuration Guide

An all-around AI assistant desktop client

Official site: https://www.cherry-ai.com/Download: https://www.cherry-ai.com/download

📋 Overview

CherryStudio is a powerful AI desktop assistant supporting major models (Claude, Gemini, GPT, and more) through a single unified interface. This guide walks you through configuring CherryStudio to use MortyCode.

✨ Features

CherryStudio offers:

• ✅ Multi-model support: Claude, Gemini, GPT, and other mainstream models
• ✅ Unified UI: Manage every AI service from one app
• ✅ Custom APIs: Plug in your own API providers
• ✅ Cross-platform: Windows, macOS, Linux
• ✅ Local-first: Data stays on your machine
• ✅ Rich functionality: Conversation management, model switching, parameter tuning, and more

🛠️ Installation

Step 1: Download and Install CherryStudio

1. Visit the CherryStudio download page

2. Pick the installer for your OS:

   • Windows: .exe installer
   • macOS: .dmg disk image
   • Linux: .AppImage or .deb package

3. Run the installer once it's downloaded.

💡 macOS Users

If you see "Can't be opened because it is from an unidentified developer", allow it via System Settings → Privacy & Security.

Step 2: Get a MortyCode API Key

Before configuring CherryStudio, grab an API Key from the MortyCode platform:

1. Visit the MortyCode console
2. Sign in
3. Create the right token group for your needs:
   • Claude models: pick the reverse-only group
   • Gemini models: pick the gemini group

4. Save the generated API Key

📋 Quick Configuration Reference

Model	Provider type	API endpoint	Token group
Claude	Anthropic	https://api.morty.live/v1/messages	Reverse-only group
Gemini	Gemini	https://api.morty.live/v1beta/models	Gemini group

🔧 Configure a Claude Model

Step 1: Open the Settings Page

1. Open CherryStudio
2. Click "Settings" or "Preferences" at the bottom left
3. Choose "Model Settings" or "API Settings"

Step 2: Choose the Claude Model Type

Pick the Claude model you want from the list.

Step 3: Configure the Claude API

Fill in the Claude configuration screen:

Parameters:

• Provider type: Anthropic
• API endpoint: https://api.morty.live
• API Key: Paste the Claude API Key from the MortyCode console
• Model name: The model you picked in the previous step (e.g. claude-sonnet-4-6)

🔧 Configure a Gemini Model

Step 1: Choose the Gemini Model Type

Pick the Gemini model you want:

• Gemini 3 Flash Preview: gemini-3-flash-preview — newest, fast and capable (recommended)
• Gemini 3 Pro Preview: gemini-3-pro-preview — high performance, great for complex tasks
• Gemini 2.0 Flash: Fast responses, good for simple chats

⚠️ Note

• Gemini and Claude require different API Keys (different token groups)
• Make sure you've created the matching token group on MortyCode

Step 2: Configure the Gemini API

Fill in the Gemini configuration screen:

Parameters:

• Provider type: Gemini
• API endpoint: https://api.morty.live/v1beta/models
• API Key: Paste the Gemini API Key from the MortyCode console
• Model name: The model you picked in the previous step (e.g. gemini-3-flash-preview)

💬 Get Started

Start a New Chat

1. Click "New Chat"
2. Select one of your configured models from the picker
3. Start chatting with the AI

Switch Models

You can change models mid-conversation:

1. Click the model picker at the top
2. Choose another configured model
3. Continue chatting (context may persist or reset depending on app settings)

Tune Parameters

CherryStudio typically supports tuning:

• Temperature: Controls randomness (0–1)
• Max Tokens: Controls response 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 chat: Temperature 0.5–0.7 (balanced)

🎯 Best Practices

1. Match Model to Task

• Code editing: 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: Gemini 3 Pro Preview (supports images)

2. Manage API Usage

• Check your balance in the MortyCode console regularly
• Use separate API Keys for separate purposes
• Don't expose API Keys in public

3. Optimize the Chat Experience

• Write clear prompts
• Keep context length reasonable
• Use the conversation history tools

🔍 Compared to Other Clients

Feature	CherryStudio	Alma	Hapi
UI type	Desktop app	Desktop app	Web/PWA
Multi-model support	✅	✅	✅
Code editing	Partial	✅	✅
Terminal integration	❌	✅	✅
Remote access	❌	❌	✅
Learning curve	Low	Medium	Medium

Recommendations:

• Chat-only needs: CherryStudio (simple, beginner-friendly)
• Development: Alma or Hapi (more capable)
• Remote control: Hapi (exclusive feature)

FAQ

"Invalid API Key" error?

Likely causes:

• API Key typo
• Wrong token group (a reverse-only Key won't work for Gemini)
• Insufficient balance

How to fix:

1. Re-copy the full API Key
2. Confirm the right token group exists in the MortyCode console (reverse-only for Claude, gemini for Gemini)
3. Top up your balance if it's low

Model list is empty?

Likely causes:

• Wrong Base URL
• Network issues
• API Key lacks permissions

How to fix:

1. Verify the Claude endpoint is https://api.morty.live/v1/messages
2. Verify the Gemini endpoint is https://api.morty.live/v1beta/models
3. Check your network
4. Regenerate the API Key and confirm the right group

Slow responses?

Likely causes:

• Network latency
• Large model selected
• Context too long

How to fix:

1. Check your network quality
2. Try a faster model (e.g. Gemini Flash)
3. Trim or clear the conversation history

How do I use multiple models at the same time?

In CherryStudio:

1. Configure each provider separately
2. Pick the desired model when starting a new chat
3. Open multiple chat windows, each with a different model

More Questions

See:

• MortyCode console
• MortyCode FAQ
• Support
• CherryStudio official docs

✅ Done

🎉 Setup complete! Enjoy chatting with AI through CherryStudio.

Key reminders:

• ✅ Different models need different API Keys (different groups)
• ✅ Claude API endpoint: https://api.morty.live/v1/messages
• ✅ Gemini API endpoint: https://api.morty.live/v1beta/models
• ✅ Match the model to the task
• ✅ Check your balance in the MortyCode console regularly
• ✅ Keep your API Keys safe

---

Related tutorials:

• Alma client configuration
• Hapi remote control
• MortyCode console