Cherry Studio
CherryHQ · AI assistants
GitHub stars
53,086Awaiting first sync
No measured star history yet
What is Cherry Studio?
Cherry Studio puts the AI models you use into one desktop app instead of a separate website for each. You bring your own API keys or run models locally, and the app keeps your conversations and knowledge bases on your computer. It suits people who switch between models every day and want chat, documents, translation and agents in one place.
Cherry Studio is a desktop AI workspace for Windows, macOS and Linux. It connects to cloud providers such as OpenAI, Anthropic and Gemini, or to local models through Ollama and LM Studio, and can compare answers from several models in one conversation. It also builds local knowledge bases, runs agents with Skills and MCP servers, and generates images.
Core capabilities
- Connect cloud providers or local models through Ollama and LM Studio in one place.
- Build local knowledge bases from files, folders, notes and web pages for chats and agents.
- Create agents with their own instructions, models and tool permissions, extended with MCP.
Many models in one window
Cherry Studio connects to OpenAI, Anthropic, Gemini and other cloud providers, plus any compatible custom service. Local models run through Ollama and LM Studio. Assistants can be customised, and a single conversation can show replies from several models. Responses render Markdown, code highlighting, formulas and Mermaid charts, and conversations can be exported.
- Multi-model
- Ollama
- LM Studio
- Assistants
Knowledge bases, agents and everyday tools
Local knowledge bases accept files such as PDF, DOCX, PPTX, XLSX, TXT and Markdown, as well as folders, notes and URLs, with keyword search by default and an optional embedding model. Agents can work with files, run commands and use browser and computer automation, and gain abilities through Skills and MCP servers. Other tools include image generation, text and document translation, web search and Markdown notes. Mini Apps open web-based AI tools inside the app, and an API Gateway lets other applications use your configured models.
- Knowledge base
- Agents
- MCP
- Translation
- Image generation
Install the desktop app
Cherry Studio is a desktop app for Windows, macOS and Linux, with separate mobile apps for iOS and Android. Getting started means adding a model provider and its API key, and cloud providers may charge for their usage. Data can be backed up locally or to WebDAV, including on a schedule. The Community Edition is licensed under AGPL-3.0, and commercial use is allowed under its terms. A commercial licence that removes the AGPL obligations, and a paid Enterprise Edition with central model management and employee access controls, are sold separately.
- AGPL-3.0
- Desktop app
- WebDAV backup
- Windows
- macOS
Where it fits
Use cases
- 01
Compare answers from several AI models
Open Settings, then Model Provider, enter an API key, select Sync models and enable the provider. In Chat, create an assistant, pick a model and add more models to the same conversation to compare their responses. Repeat with other providers or a local model in Ollama.
- 02
Ask questions about your own documents
Open Knowledge Base, click New Knowledge Base and add files, notes, folders or URLs with Add Data Source. Wait until the items show as ready, then run a recall test with questions you already know the answers to. Select the knowledge base in the chat input, or bind it to an agent for ongoing work.
- 03
Hand repeat tasks to an agent
Create an agent with its own instructions, workspace, model and tool permissions, and add Skills or MCP servers to extend what it can do. The agent can work with files, run commands and use browser automation tools. Scheduled tasks run it on a timetable, and Channels connect it to messaging platforms.