📌Overview

Purpose:
To provide a fast, lightweight, and privacy-focused AI assistant supporting multiple leading large language models, customizable for individuals and enterprises.

Overview:
NextChat is a cross-platform AI assistant framework that integrates top AI models such as Claude, DeepSeek, GPT-4, and Gemini Pro. It is designed for rapid deployment and ease of use, offering free self-hosting options, robust privacy protections, and extensive customization, including enterprise-grade features. Its lightweight client and broad compatibility make it ideal for both personal productivity and organizational deployment.

Key Features:

Multi-Model Support:
Seamlessly integrates with popular AI models (Claude, GPT-4, Gemini Pro, DeepSeek, and more), as well as self-hosted solutions like RWKV-Runner and LocalAI, allowing users to choose the best LLM for their needs.
Privacy & Customization:
Stores all user data locally to prioritize privacy and offers extensive customization options, including custom prompts, brand adaptation, fine-grained permission controls, and enterprise-specific deployments.

NextChat

Lightweight, fast AI Assistant with support for Claude, DeepSeek, GPT-4, and Gemini Pro.

Available at:

🥳 NextChat iOS Version is Online!

Install on iOS

Enterprise Edition

NextChat supports enterprise privatization and customization:

Brand Customization: Tailor the UI to match your corporate identity.
Resource Integration: Centralized management of multiple AI resources.
Permission Control: Granular admin-managed permissions.
Knowledge Integration: Use company-internal knowledge with AI.
Security Auditing: Sensitive inquiry interception and full conversation history traceability.
Private Deployment: Supports private cloud deployment for security.
Continuous Updates: Regular capability upgrades, including multimodal AI.

For enterprise inquiries, contact: business@nextchat.dev

Features

One-click free deployment via Vercel in under 1 minute.
Compact client (~5MB) for Linux/Windows/MacOS: Download
Compatible with self-hosted LLMs, recommended with RWKV-Runner or LocalAI
Privacy first: all data stored locally in your browser.
Markdown support: LaTeX, mermaid, code highlighting, etc.
Responsive design, dark mode, PWA.
Quick first load (~100kb), streaming responses.
Prompt templates: create, share, and debug chat tools.
Built-in awesome prompts (awesome-chatgpt-prompts-zh, awesome-chatgpt-prompts)
Automatic chat history compression for long conversations.
I18n: English, 简体中文, 繁体中文, 日本語, Français, Español, Italiano, Türkçe, Deutsch, Tiếng Việt, Русский, Čeština, 한국어, Indonesia.

Roadmap

What's New

v2.15.8: Realtime Chat support
v2.15.4: Tauri fetch support for LLM API
v2.15.0: Plugin support (NextChat-Awesome-Plugins)
v2.14.0: Artifacts & Stable Diffusion
v2.10.1: Google Gemini Pro model support
v2.9.11: Azure endpoint support
v2.8: Cross-platform client
v2.7: Share conversations as image
v2.0: Prompt templates

Get Started

Get your OpenAI API Key
Deploy with Vercel:
Enjoy!

Keeping Updated

If deploying by Vercel, to ensure updates work:

Fork this repository.
Deploy from your fork on Vercel.
Enable GitHub Actions for workflows and upstream sync.
To update instantly, see GitHub documentation.

Access Password

To restrict access, set the CODE environment variable in Vercel to a comma-separated list of passwords:

code1,code2,code3

Redeploy after changes.

Environment Variables

Key variables for configuration:

CODE (optional): Comma-separated access passwords.
OPENAI_API_KEY (required): Your OpenAI API key(s), comma-separated.
BASE_URL (optional): Override OpenAI API request base URL.
OPENAI_ORG_ID (optional): OpenAI organization ID.
AZURE_URL, AZURE_API_KEY, AZURE_API_VERSION: Azure support.
GOOGLE_API_KEY, GOOGLE_URL: Google Gemini Pro support.
ANTHROPIC_API_KEY, ANTHROPIC_URL, ANTHROPIC_API_VERSION: Anthropic Claude support.
BAIDU_API_KEY, BAIDU_SECRET_KEY, BAIDU_URL: Baidu support.
BYTEDANCE_API_KEY, BYTEDANCE_URL: ByteDance support.
ALIBABA_API_KEY, ALIBABA_URL: Alibaba Cloud support.
IFLYTEK_URL, IFLYTEK_API_KEY, IFLYTEK_API_SECRET: iFlytek support.
CHATGLM_API_KEY, CHATGLM_URL: ChatGLM support.
DEEPSEEK_API_KEY, DEEPSEEK_URL: DeepSeek support.
HIDE_USER_API_KEY: Set to 1 to prevent user-key input.
DISABLE_GPT4: Set to 1 to disable GPT-4.
ENABLE_BALANCE_QUERY
DISABLE_FAST_LINK
CUSTOM_MODELS: Control which models are shown, using ± prefixes.
DEFAULT_MODEL
VISION_MODELS
WHITE_WEBDAV_ENDPOINTS
DEFAULT_INPUT_TEMPLATE
STABILITY_API_KEY, STABILITY_URL
ENABLE_MCP: Enable Model Context Protocol feature.
SILICONFLOW_API_KEY, SILICONFLOW_URL

Requirements

NodeJS >= 18
Docker >= 20

Development

Before starting development, create a .env.local file in the project root and add your API key:

OPENAI_API_KEY=<your api key here>
BASE_URL=https://chatgpt1.nextweb.fun/api/proxy  # if OpenAI is unavailable

Run locally:

yarn install
yarn dev

Deployment

Docker (Recommended):

docker pull yidadaa/chatgpt-next-web

docker run -d -p 3000:3000 \
   -e OPENAI_API_KEY=sk-xxxx \
   -e CODE=your-password \
   yidadaa/chatgpt-next-web

If using a proxy:

docker run -d -p 3000:3000 \
   -e OPENAI_API_KEY=sk-xxxx \
   -e CODE=your-password \
   -e PROXY_URL=http://localhost:7890 \
   yidadaa/chatgpt-next-web

Enable MCP:

docker run -d -p 3000:3000 \
   -e OPENAI_API_KEY=sk-xxxx \
   -e CODE=your-password \
   -e ENABLE_MCP=true \
   yidadaa/chatgpt-next-web

Shell Script:

bash <(curl -s https://raw.githubusercontent.com/Yidadaa/ChatGPT-Next-Web/main/scripts/setup.sh)

Synchronizing Chat Records (UpStash)

Documentation for syncing chat records is available in multiple languages, including English, Chinese, Italian, Japanese, and Korean.

Documentation

See the docs directory for more information, such as:

Deploying with Cloudflare
FAQ
Adding translations
Using Vercel
User manual (Chinese, WIP)

Translation

To add a new translation, refer to the translation documentation.

Donation

Buy Me a Coffee

License

MIT