Repolode
Submit
ChatGPTNextWeb

ChatGPTNextWeb/NextChat

Open-source, cross-platform AI chat client for Web, desktop and mobile that connects to OpenAI, Claude, Gemini, DeepSeek and self-hosted models.

Open-source alternative to:ChatGPTClaude

View on GitHubWebsite
Stars
88,874
Forks
58,766
Open issues
853
Last commit
11 hours ago

What is it?

NextChat is a lightweight AI chat interface built with TypeScript and Next.js. It runs as a web app, a PWA, and native desktop and iOS builds, and supports models from OpenAI, Anthropic Claude, Google Gemini, DeepSeek and others, as well as self-hosted LLMs through tools like Ollama, RWKV-Runner and LocalAI.

The project can be deployed in one click to Vercel or run in Docker, and access can be restricted with a password set through environment variables. Chat data is stored locally in the browser by default. Users can create prompt templates called masks, share conversations as images, and use Markdown with LaTeX, Mermaid diagrams and code highlighting.

Recent versions add MCP support (enabled by an environment flag), plugins, Artifacts for previewing generated pages, and realtime chat. An enterprise edition with admin controls and private deployment is offered separately.

Key features

  • 01One-click deployment to Vercel or Docker
  • 02Connects to multiple cloud models and self-hosted LLMs
  • 03Local browser storage of chat data
  • 04Prompt templates (masks) for reusable chat setups
  • 05Markdown rendering with LaTeX, Mermaid and code highlighting
  • 06Interface available in many languages with dark mode and PWA support

Use cases

  • Running a private, password-protected chat front end for a team using API keys
  • Chatting with locally hosted models through Ollama or LocalAI
  • Using one interface across desktop, mobile and web for multiple AI providers

Quick start

  1. 1

    Get an OpenAI API key from the OpenAI platform

  2. 2

    Click the Deploy with Vercel button and set CODE as your page password

  3. 3

    Open the deployed site and start chatting

Things to know

  • Requires an API key from a model provider such as OpenAI, which is billed separately.
  • Deploying your own copy needs Vercel or a similar hosting setup and environment variable configuration.

Topics

#chatbot#llm#nextjs#typescript#self-hosted#openai#claude#gemini#calclaude#cross-platform

FAQ

What is NextChat?
Open-source, cross-platform AI chat client for Web, desktop and mobile that connects to OpenAI, Claude, Gemini, DeepSeek and self-hosted models. NextChat is a lightweight AI chat interface built with TypeScript and Next.js. It runs as a web app, a PWA, and native desktop and iOS builds, and supports models from OpenAI, Anthropic Claude, Google Gemini, DeepSeek and others, as well as self-hosted LLMs through tools like Ollama, RWKV-Runner and LocalAI.
Is NextChat free to use? What license does it have?
NextChat is open source under the MIT license. Check the license terms before using it in commercial products.
Is NextChat still maintained?
Actively maintained. The last commit to NextChat was on Oct 20, 2026.
What can I use NextChat for?
Running a private, password-protected chat front end for a team using API keys Chatting with locally hosted models through Ollama or LocalAI Using one interface across desktop, mobile and web for multiple AI providers
How do I install NextChat?
Get an OpenAI API key from the OpenAI platform Click the Deploy with Vercel button and set CODE as your page password Open the deployed site and start chatting
Where can I get NextChat?
The source code and installation instructions are on GitHub: https://github.com/ChatGPTNextWeb/NextChat

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects