chat-js

ChatJS is an open‑source starter kit for building AI chat applications. It bundles authentication, streaming UI, tool calling and access to more than 120 models from providers such as OpenAI, Anthropic, Google, xAI and Meta. The project is released under the Apache 2.0 license and is positioned as a production‑ready foundation rather than a simple demo.

The kit runs on a modern Next.js stack with TypeScript, Tailwind CSS and a collection of libraries that support type safety and reliability. You can launch a web version instantly from the live demo, or follow the desktop guide to install a native app for macOS, Windows or Linux. The builds are unsigned, so the installation process includes steps to bypass OS warnings. For developers who prefer a command‑line start, the CLI creates a new project with a single command (`npx @chat-js/cli@latest create my-app`), after which you can customize the code, add your own branding and deploy to your preferred host.

ChatJS is aimed at developers and product teams that need a quick baseline for AI copilots, internal assistants, customer‑facing chat products or multi‑model playgrounds. It supports parallel responses, letting a single user message be sent to multiple models at once so you can compare quality, tone and accuracy side‑by‑side. Built‑in tools include a multi‑step research agent that pulls web data into reports, a code execution sandbox for Python and JavaScript (with pandas, numpy, matplotlib pre‑installed), real‑time web search with inline citations, and image or video generation directly within a conversation. The platform also offers a “Canvas” for editing rich documents, code files and spreadsheets with version history, inline diffs and collaborative editing.

Authentication is handled out of the box with OAuth providers Google, GitHub and Vercel, plus session management, anonymous mode and rate limiting. The Model Context Protocol (MCP) lets you extend capabilities with external services, while “Branching” allows you to fork a conversation and explore alternative paths without losing context. All of these features are designed to reduce the months‑long effort typically required to build a production chat backend.

Key features

• Support for 120+ AI models from major providers through a single interface
• Built‑in OAuth authentication (Google, GitHub, Vercel) with session handling
• Parallel model responses and side‑by‑side comparison
• Canvas for collaborative editing of documents, code and spreadsheets