SearChat

Overview of SearChat

What it is

SearChat is an AI-powered conversational search engine built with a Turborepo monorepo architecture, integrating a Node.js + Koa backend and a Vue 3 + TypeScript frontend. It connects to multiple AI model providers (OpenAI, Anthropic, Google Gemini, Vertex AI) and supports various search engines, including SearXNG, Bing, Google, Tavily, Exa, Bocha, and ChatGLM Web Search. The project is open source under the MIT license and provides a Docker-based one-click deployment option.

Key features

• Multi-model support – Compatible with OpenAI, Anthropic, Gemini, and Vertex AI APIs.
• Multiple search engines – Supports Bing, Google, SearXNG, Tavily, Exa, Bocha, and ChatGLM Web Search.
• Conversational search – Multi-turn chat-based search experience with history cached in IndexedDB/LocalStorage.
• Deep Research mode – AI-driven iterative search and structured report generation using LangChain + LangGraph.
• Modern interface – Responsive design, dark/light theme, i18n, real-time streaming (typewriter effect), and contextual multi-turn dialogue.
• Docker deployment – One-click setup via docker-compose with environment variables for API keys and SearXNG configuration.

Use cases

SearChat is suitable for users who need an AI-powered search interface that can aggregate results from multiple sources and support deep research. It can be used for exploratory research, generating structured reports, or as a private conversational search tool when deployed with SearXNG. Developers can also integrate the Deep Research module into their own Node.js projects via the deepsearcher npm package.

Getting started

To deploy SearChat with Docker:bash# Create docker-compose.yaml and model.json as described in the READMEdocker compose up -d# Open http://localhost:3000For detailed configuration steps, see the repository README.

License and pricing

The project is licensed under the MIT license. Pricing is not stated; it is open source.

這是什麼

SearChat 是一個由 AI 驅動的對話式搜尋引擎，採用 Turborepo 單體式架構，整合 Node.js + Koa 後端與 Vue 3 + TypeScript 前端。它連接多個 AI 模型提供商（OpenAI、Anthropic、Google Gemini、Vertex AI），並支援多種搜尋引擎，包括 SearXNG、Bing、Google、Tavily、Exa、Bocha 與 ChatGLM 網頁搜尋。本專案採用 MIT 授權，並提供基於 Docker 的一鍵部署方式。

主要功能

• 多模型支援 – 相容 OpenAI、Anthropic、Gemini 與 Vertex AI API。
• 多搜尋引擎 – 支援 Bing、Google、SearXNG、Tavily、Exa、Bocha 與 ChatGLM 網頁搜尋。
• 對話式搜尋 – 多輪聊天式搜尋體驗，對話歷史會快取於 IndexedDB/LocalStorage。
• 深度研究模式 – 使用 LangChain + LangGraph 進行 AI 驅動的迭代搜尋與結構化報告生成。
• 現代化介面 – 響應式設計、深色/淺色主題、多國語言、即時串流（打字機效果）與上下文多輪對話。
• Docker 部署 – 透過 docker-compose 一鍵設定，可配置環境變數設定 API 金鑰與 SearXNG。

使用情境

SearChat 適合需要由 AI 驅動、可彙整多個來源搜尋結果並支援深度研究的用戶。可用於探索性研究、生成結構化報告，或透過 SearXNG 部署作為私有的對話式搜尋工具。開發者亦可將深度研究模組以 deepsearcher npm 套件整合至自己的 Node.js 專案。

快速開始

使用 Docker 部署 SearChat：bash# 建立 docker-compose.yaml 與 model.json（詳細步驟請見 README）docker compose up -d# 開啟 http://localhost:3000詳細設定步驟請參閱專案 README。

授權與價格

本專案採用 MIT 授權。價格並未提及；此為開源軟體。

Based on the project's public README and repository metadata.