Cherry Studio

About Cherry Studio

Open-source desktop AI client that puts local runtimes, remote provider APIs, MCP servers, and document knowledge bases behind one chat interface.

Best for: People who want one desktop chat client for both local and hosted models, and who want MCP tool access without wiring up a separate agent framework.

Deployment: Native desktop application for macOS, Windows, and Linux; AGPL-3.0 Electron app that talks to local runtimes such as Ollama and LM Studio alongside remote provider APIs

Skill level: Beginner

Tradeoffs

The AGPL-3.0 copyleft is a real constraint if you intend to modify Cherry Studio and offer it to others over a network — that triggers a source-disclosure obligation, and the project sells a commercial licence specifically as an exemption from it. Team features such as centrally managed model access, shared knowledge bases, and role-based permissions belong to a separate paid Enterprise edition, not the AGPL community edition. Licensing has also moved before: the project previously shipped under Apache 2.0 with added commercial-use restrictions and only relicensed to clean AGPL-3.0 during 2025, so verify the licence yourself if you are planning a long-lived commercial deployment.

Related guides and resources

Explore step-by-step setup guides, comparisons, and stack recipes for this tool category.

Best for

People who want one desktop chat client for both local and hosted models, and who want MCP tool access without wiring up a separate agent framework.

Why use it

Cherry Studio is one of the few open-source desktop chat clients with first-class MCP (Model Context Protocol) support built in, so tools and data sources can be attached to a conversation without writing an agent harness. It also treats local runtimes and hosted provider APIs as peers, which makes it a practical front end while you are still deciding how much of your workload should run locally. Document knowledge bases are handled in-app, so basic retrieval over your own files does not require standing up a separate RAG stack.

Key features

• Built-in MCP (Model Context Protocol) server support, letting a conversation reach external tools and data sources without a separate agent framework
• Single interface over both local runtimes and remote provider APIs, so local and hosted models can be compared side by side
• In-app document knowledge bases for retrieval over your own files without deploying a separate RAG stack
• Cross-platform Electron desktop app for macOS, Windows, and Linux with an auditable AGPL-3.0 codebase

Tradeoffs

The AGPL-3.0 copyleft is a real constraint if you intend to modify Cherry Studio and offer it to others over a network — that triggers a source-disclosure obligation, and the project sells a commercial licence specifically as an exemption from it. Team features such as centrally managed model access, shared knowledge bases, and role-based permissions belong to a separate paid Enterprise edition, not the AGPL community edition. Licensing has also moved before: the project previously shipped under Apache 2.0 with added commercial-use restrictions and only relicensed to clean AGPL-3.0 during 2025, so verify the licence yourself if you are planning a long-lived commercial deployment.

Alternatives

• Jan
• Chatbox
• AnythingLLM