Cherry Studio
A cross-platform desktop workspace for using cloud and local language models from one client.
Per-dimension scores and reasoning
Evidence shows a security policy and dependency scanning claims, but no implementation details. User confirmation mechanisms are not documented, data flow transparency is limited, sensitive data handling is not detailed. Dependency security is claimed but lacks evidence, external effects (e.g., network requests) are not explicit, rollback is not mentioned, source attribution is partial (e.g., author email). Thus scores are low.
Project has test files and CI configuration, indicating some consistency. Dependency list is complete but availability not verified. Failure messages are not detailed in documentation.
README clearly targets desktop users, supports multiple platforms, broad scenarios. Capability boundaries are not explicit, trigger precision not addressed, environment fit is documented (cross-platform).
Information architecture is clear, with README, docs links, contribution guide. Install notes are provided (e.g., pnpm install). Naming is stable (version number explicit). Examples and FAQ are sparse, known limitations not explicit, license is clear (AGPL-3.0), versioning changelog is mentioned (changeset), maintenance responsibility is community and contribution guide.
Output usability is high (desktop app), marginal value is high (multi-model support), cost-benefit is not explicit (commercial edition exists but no pricing).
Claims are partially traceable (e.g., feature list), but lack independent verification. Cross-source corroboration is limited, facts and inferences are not clearly separated.
- Dependency security is not verified; check for vulnerabilities.
- User confirmation mechanisms are not explicit; verify sensitive operations.
- Data flow transparency is limited; review network requests and data storage.
What does this agent do, and when should you use it?
Cherry Studio is a desktop client available for Windows, Mac, and Linux. It brings together cloud services including OpenAI, Gemini, and Anthropic with local-model options through Ollama and LM Studio. The product includes more than 300 preconfigured AI assistants, custom assistant creation, and simultaneous conversations with multiple models. Its documented workspace features also cover text, images, Office files, PDFs, Markdown rendering, Mermaid charts, code highlighting, search, and topic management. The Community Edition is licensed under AGPL-3.0; the repository also describes a privately deployable Enterprise Edition whose features are only partially released to customers.
In the desktop client, a user selects a supported cloud LLM service or uses a local model through Ollama or LM Studio, then works through a preconfigured or custom AI assistant. It supports simultaneous conversations with multiple models and accepts text, images, Office files, and PDFs as documented content types. The client renders Markdown, visualizes Mermaid charts, highlights code syntax, and provides AI-powered translation, global search, themes, and topic management. WebDAV is listed for file management and backup, while MCP (Model Context Protocol) Server is listed among the integrated tools.
- An individual who wants one Windows, Mac, or Linux client for OpenAI, Gemini, Anthropic, and local models.
- A researcher or content professional who wants to ask several models the same question and compare their responses.
- A desktop user who needs AI conversations to work with PDFs, Office documents, images, or text materials.
- A developer or privacy-conscious user who already runs Ollama or LM Studio and wants a graphical client for local models.
- A knowledge worker who wants WebDAV-based file management and backup for the application.
- A team member who wants to start with preconfigured assistants while retaining the ability to create specialized ones.
What are this agent's strengths and limitations?
- One desktop client spans cloud services such as OpenAI, Gemini, and Anthropic as well as local models through Ollama and LM Studio.
- Simultaneous multi-model conversations support direct response comparison for the same task.
- The documented workflow extends beyond chat to text, images, Office files, PDFs, Markdown, Mermaid, and code presentation.
- It is explicitly available across Windows, Mac, and Linux and includes WebDAV file management and backup.
- The README provides no copyable installation commands, system requirements, model-connection configuration, or first-run walkthrough.
- Cloud-model use depends on the relevant services; the supplied material does not document key handling, pricing, offline boundaries, or provider-specific differences.
- The Community Edition uses AGPL-3.0, so commercial use must comply with that license; an exemption requires a separate commercial license.
- Private deployment, centralized management, and access controls are described for an Enterprise Edition that is only partially released to customers, so they should not be assumed for the Community Edition.
How do you install or deploy this agent?
The README states that Cherry Studio is available for Windows, Mac, and Linux and links to GitHub Releases, but it supplies no copyable download or installation command, package name, minimum system requirements, or installation procedure. It also omits the credential setup for cloud services and the connection steps for Ollama or LM Studio needed for a first working model.
How do you use this agent?
The documented usage path is to choose a cloud LLM service in the desktop client or use a local model through Ollama or LM Studio, then select one of the 300+ preconfigured assistants or create a custom assistant and begin a conversation or multi-model conversation. Users can work with text, images, Office files, and PDFs, and use Markdown, Mermaid, code highlighting, translation, search, and topic-management features. The supplied README does not provide UI steps, commands, or configuration fields for connecting the first model or sending the first prompt.
How does this agent compare with similar options?
The README positions Cherry Studio as a desktop client that unifies services including OpenAI, Gemini, Anthropic, Claude, Perplexity, Poe, Ollama, and LM Studio. It lists new-api and one-api as related projects for LLM gateway, API-management, or API-distribution use cases, which is a different role from a desktop chat client.