Chatbox
Desktop client for ChatGPT, Claude, Gemini, and local models with all data stored on your machine.
About
A bring-your-own-key desktop client is Chatbox's core proposition: install a package on Windows 10 or later, macOS 11 or later, or Linux, paste API keys, and chat with OpenAI, Azure OpenAI, Claude, and Google Gemini, or run local models through Ollama, with iOS, Android, and web versions extending the same interface. All conversation data is stored locally rather than on a vendor server, and the feature set covers Markdown and LaTeX rendering, code syntax highlighting, a prompt library, message quoting, keyboard shortcuts, streaming responses, dark theme, DALL-E 3 image generation, and a localized interface in many languages. The community edition on GitHub is GPL-3.0 and free, while a separate Chatbox AI service offers hosted models for users without their own keys; the maintainers regularly sync code between the community and pro repositories. No GPU or technical setup is required beyond running an installer. With 41.2k GitHub stars it is one of the longest-standing and most popular desktop LLM clients across all three major operating systems.
Should you use Chatbox?
Pick it when
Pick Chatbox when one person wants a plain desktop or phone app to use their own OpenAI, Claude, or Gemini keys, or a local Ollama model, with chats stored on the device and nothing to host or configure.
Look elsewhere when
Skip it if you want a deeper toolkit, such as Cherry Studio's MCP marketplace and 300+ assistants, or a shared multi-user server like Open WebUI or LibreChat. GPL-3.0 also matters if you plan to ship a modified fork.
Alternatives to Chatbox
- Cherry Studio
Adds an MCP marketplace, 300+ ready assistants, and WebDAV backup on the desktop; Chatbox stays the pick when you also want official iOS and Android apps.
- NextChat
MIT licensed with a tiny Tauri desktop app and one-click Vercel deploy, a better pick if you also want a self-hosted personal web instance.
- big-AGI
Browser-based with Beam to compare and merge several models' answers, better for cross-checking output, but a web app rather than a desktop installer.
- Jan
Bundles its own llama.cpp engine and GGUF model downloads, so fully offline chat needs no separate Ollama install as Chatbox does, and it can serve models on a local API.
Reviews (0)
Leave a Review
No reviews yet. Be the first to review!
Details
- Category
- UI & Chat Interfaces
- Price
- Freemium
- Platform
- Hybrid
- Difficulty
- Beginner (1/5)
- License
- GPL-3.0
- Added
- Jul 29, 2026