Skip to content

yhemeo613/smartchat

BranchesTags

Repository files navigation

SmartChat

Turn your knowledge base into an intelligent AI customer support agent.

online

SmartChat

将你的知识库变成智能 AI 客服。

Next.js Supabase TypeScript License

English · 中文

What is SmartChat?

SmartChat is a free, open-source AI chatbot platform. Upload your documents, train a chatbot in minutes, and embed it on any website with one line of code. It supports multiple LLM providers — OpenAI, Anthropic, DeepSeek, Qwen, Zhipu GLM, Moonshot, or any OpenAI-compatible API.

Screenshots

Homepage Features
Homepage Features
Dashboard Bot Settings
Dashboard Bot Settings
Knowledge Base Chat
Knowledge Base Chat
Embed Code Settings
Embed Settings
Demo
Demo

Features

  • RAG-Powered Answers — Upload PDF, TXT, DOCX, MD files. The retrieval-augmented generation engine finds the most relevant answers from your knowledge base.
  • Multi-Provider LLM — Switch between OpenAI, Anthropic, DeepSeek, Qwen, Zhipu, Moonshot, or bring your own OpenAI-compatible endpoint.
  • One-Click Embed — Add a chat widget to any website with a single <script> tag.
  • Fully Customizable — Custom colors, avatars, welcome messages, system prompts, temperature, and model selection per bot.
  • Real-Time Streaming — SSE-based streaming responses for a natural chat experience.
  • Multilingual — UI supports English and Chinese. The AI responds in the visitor's language.
  • Local Embeddings — Built-in support for local embedding models (BGE-small-zh, all-MiniLM-L6-v2) — no API key needed for vectorization.
  • Conversation Analytics — Track conversations, visitor sessions, and message history.

Tech Stack

Layer Technology
Framework Next.js 16 (App Router, Turbopack)
Language TypeScript 5
Database Supabase (PostgreSQL + pgvector)
Auth Supabase Auth
AI SDKs OpenAI SDK, Anthropic SDK
Embeddings @huggingface/transformers (local)
UI Tailwind CSS 4, Radix UI, shadcn/ui
Animation Framer Motion

Project Structure

src/
├── app/
│   ├── (dashboard)/     # Dashboard pages (bots, settings)
│   ├── (marketing)/     # Landing page, docs, donate
│   ├── api/             # API routes (chat, upload, bots, user settings)
│   ├── chat/[botId]/    # Public chat page
│   └── demo/            # Demo page
├── components/
│   ├── dashboard/       # Bot management, document upload, embed code
│   ├── marketing/       # Hero, features, navbar, footer
│   └── ui/              # shadcn/ui components
└── lib/
    ├── ai/              # LLM provider routing, RAG, embeddings, provider presets
    ├── i18n/            # Internationalization (en/zh)
    └── supabase/        # Client, server, service client, schema

Getting Started

Prerequisites

  • Node.js 18+
  • A Supabase project (free tier works)

1. Clone & Install

git clone https://github.com/yhemeo613/smartchat.git
cd smartchat
npm install

2. Configure Environment

Create .env.local:

NEXT_PUBLIC_SUPABASE_URL=https://your-project.supabase.co
NEXT_PUBLIC_SUPABASE_ANON_KEY=your_anon_key
SUPABASE_SERVICE_ROLE_KEY=your_service_role_key
NEXT_PUBLIC_APP_URL=http://localhost:3000

3. Set Up Database

Open the Supabase SQL Editor and run the contents of src/lib/supabase/schema.sql. This creates all tables, indexes, RLS policies, and the vector similarity search function.

4. Run

npm run dev

Open http://localhost:3000.

AI Provider Configuration

Go to Dashboard → Settings to configure your AI provider. Supported providers:

Provider Base URL Models
OpenAI https://api.openai.com/v1 GPT-4o, GPT-4o Mini, GPT-3.5 Turbo
Anthropic https://api.anthropic.com Claude Sonnet 4, Claude 3.5 Haiku
DeepSeek https://api.deepseek.com/v1 DeepSeek Chat (V3), DeepSeek Reasoner (R1)
Qwen (通义千问) https://dashscope.aliyuncs.com/compatible-mode/v1 Qwen Max, Qwen Plus, Qwen Turbo
Zhipu GLM (智谱) https://open.bigmodel.cn/api/paas/v4 GLM-4 Plus, GLM-4 Flash
Moonshot (月之暗面) https://api.moonshot.cn/v1 Moonshot v1 8K/32K/128K
Custom Any OpenAI-compatible URL Any model name

Embed on Your Website

After creating a bot, copy the embed code from the Embed tab:

<script src="https://your-domain.com/widget.js" data-bot-id="your-bot-id"></script>

Deploy

Deploy to Vercel with one click:

Deploy with Vercel

Or build and run manually:

npm run build
npm start

Support the Project

If you find SmartChat helpful, consider buying the author a coffee :)

WeChat Pay Alipay
WeChat Pay Alipay

Contributing

Contributions are welcome! Feel free to open issues and pull requests.

Author

豆豆 — dzyperson@163.com

License

MIT

SmartChat 是什么？

SmartChat 是一个免费开源的 AI 聊天机器人平台。上传文档，几分钟内训练聊天机器人，一行代码嵌入任何网站。支持多家 LLM 服务商 —— OpenAI、Anthropic、DeepSeek、通义千问、智谱 GLM、月之暗面，或任何兼容 OpenAI API 的服务。

界面预览

首页 功能特性
首页 功能特性
控制台 机器人设置
控制台 机器人设置
知识库 聊天
知识库 聊天
嵌入代码 设置
嵌入代码 设置
演示
演示

功能特性

  • RAG 智能问答 — 上传 PDF、TXT、DOCX、MD 文件，检索增强生成引擎从知识库中找到最相关的答案。
  • 多厂商 LLM — 一键切换 OpenAI、Anthropic、DeepSeek、通义千问、智谱、月之暗面，或自定义 OpenAI 兼容接口。
  • 一键嵌入 — 一个 <script> 标签即可在任何网站添加聊天组件。
  • 完全可定制 — 自定义颜色、头像、欢迎语、系统提示词、创造性参数，每个机器人独立配置模型。
  • 实时流式响应 — 基于 SSE 的流式输出，带来自然的聊天体验。
  • 中英双语 — 界面支持中文和英文，AI 自动使用访客的语言回复。
  • 本地向量化 — 内置本地 Embedding 模型（BGE-small-zh、all-MiniLM-L6-v2），向量化无需 API Key。
  • 对话分析 — 追踪对话记录、访客会话和消息历史。

技术栈

层级 技术
框架 Next.js 16 (App Router, Turbopack)
语言 TypeScript 5
数据库 Supabase (PostgreSQL + pgvector)
认证 Supabase Auth
AI SDK OpenAI SDK, Anthropic SDK
向量化 @huggingface/transformers（本地）
UI Tailwind CSS 4, Radix UI, shadcn/ui
动画 Framer Motion

快速开始

前置要求

  • Node.js 18+
  • 一个 Supabase 项目（免费版即可）

1. 克隆 & 安装

git clone https://github.com/yhemeo613/smartchat.git
cd smartchat
npm install

2. 配置环境变量

创建 .env.local：

NEXT_PUBLIC_SUPABASE_URL=https://your-project.supabase.co
NEXT_PUBLIC_SUPABASE_ANON_KEY=your_anon_key
SUPABASE_SERVICE_ROLE_KEY=your_service_role_key
NEXT_PUBLIC_APP_URL=http://localhost:3000

3. 初始化数据库

打开 Supabase SQL Editor，运行 src/lib/supabase/schema.sql 的内容。这会创建所有表、索引、RLS 策略和向量相似度搜索函数。

4. 启动

npm run dev

打开 http://localhost:3000。

AI 服务商配置

进入 控制台 → 设置 配置 AI 服务商。支持的服务商：

服务商 Base URL 模型
OpenAI https://api.openai.com/v1 GPT-4o, GPT-4o Mini, GPT-3.5 Turbo
Anthropic https://api.anthropic.com Claude Sonnet 4, Claude 3.5 Haiku
DeepSeek https://api.deepseek.com/v1 DeepSeek Chat (V3), DeepSeek Reasoner (R1)
通义千问 https://dashscope.aliyuncs.com/compatible-mode/v1 Qwen Max, Qwen Plus, Qwen Turbo
智谱 GLM https://open.bigmodel.cn/api/paas/v4 GLM-4 Plus, GLM-4 Flash
月之暗面 https://api.moonshot.cn/v1 Moonshot v1 8K/32K/128K
自定义 任何 OpenAI 兼容 URL 任意模型名

嵌入到你的网站

创建机器人后，从 嵌入 标签页复制代码：

<script src="https://your-domain.com/widget.js" data-bot-id="your-bot-id"></script>

部署

一键部署到 Vercel：

Deploy with Vercel

或手动构建运行：

npm run build
npm start

打赏支持

SmartChat 完全免费开源。如果觉得有帮助，可以请作者喝杯咖啡 :)

微信支付 支付宝
微信支付 支付宝

贡献

欢迎贡献！请随时提交 Issue 和 Pull Request。

作者

豆豆 — dzyperson@163.com

开源协议

MIT

About

smartchat-lovat.vercel.app

Resources

Readme
Activity

Stars

1 star

Watchers

0 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

 
 
 

Contributors

Languages