Chatbox

About Chatbox

Cross-platform desktop AI client that talks to local runtimes and cloud providers through one interface, with a built-in Ollama provider, document knowledge bases, and local storage of conversations.

Best for: People who want a polished desktop chat application over their local models — a single window for Ollama, LM Studio and cloud providers, without running a web server or living in a browser tab.

Deployment: Prebuilt desktop application for Windows, macOS and Linux from the project's releases page — a Linux AppImage is provided. Building from source requires Node 22.x and pnpm 10+; on Linux the Electron sandbox helper must be owned by root with mode 4755 before the app will start.

Skill level: Beginner

Tradeoffs

GPL-3.0 is a strong copyleft licence — fine for personal and internal use, but check compatibility before bundling or redistributing it as part of a product. It is a desktop GUI, so it needs a graphical session and is not suitable for a headless server. The application checks a vendor endpoint for built-in skills at startup, so a fully air-gapped install needs that path reviewed.

Related guides and resources

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

Best for

People who want a polished desktop chat application over their local models — a single window for Ollama, LM Studio and cloud providers, without running a web server or living in a browser tab.

Why use it

Chatbox is a native desktop application rather than a self-hosted web app, so there is no container to run or port to manage — install it and point it at a provider. Ollama is a defined provider in its source with a default endpoint of http://127.0.0.1:11434, wired through the same interface as every cloud backend, so local models are a first-class option rather than a bolt-on. Conversations and knowledge-base data are stored locally on your machine. Note the GPL-3.0 licence if you intend to build on or redistribute it.

Key features

• Built-in Ollama provider defaulting to 127.0.0.1:11434, alongside vLLM, LM Studio and cloud backends in one client
• Local document knowledge base with retrieval over your own files
• Conversations stored on the local machine rather than a vendor account
• Cross-platform desktop builds for Windows, macOS and Linux
• Artifact and image handling with a session-scoped attachment store

Common AI use cases

• Chat with local Ollama models from a desktop app instead of a terminal or browser tab
• Keep one client configured for both a local model and a cloud provider, switching per conversation
• Query your own documents through a local knowledge base

Who should use it

• Beginners who want a working local chat interface without self-hosting a web UI
• Anyone switching regularly between local and cloud models who wants one window for both
• Users who want conversation history stored locally rather than in a vendor account

Who should not use it

• Teams needing a shared, multi-user server-hosted interface — a self-hosted web UI fits better
• Projects that cannot accept GPL-3.0 terms for redistribution
• Headless servers with no graphical session

Tradeoffs

GPL-3.0 is a strong copyleft licence — fine for personal and internal use, but check compatibility before bundling or redistributing it as part of a product. It is a desktop GUI, so it needs a graphical session and is not suitable for a headless server. The application checks a vendor endpoint for built-in skills at startup, so a fully air-gapped install needs that path reviewed.

Alternatives

• Jan
• LM Studio