一、下載LangChain chat ui

下載地址：點(diǎn)此下載 ，也可以使用官網(wǎng)線上版本 地址：https://agentchat.vercel.app/

二、安裝LangGraph cli

pip install -U langgraph
pip install -U "langgraph-cli[inmem]"

三、項(xiàng)目?jī)?nèi)創(chuàng)建langgraph.json

{
  "dependencies": ["."],
  "graphs": {
    "agent": "./supervisor_agent.py:supervisor_agent"  // agent的python文件：agent的名稱
  },
  "env": ".env"
}

四、執(zhí)行啟動(dòng)命令

langgraph dev --host 0.0.0.0  # 看到啟動(dòng)界面如下
INFO:langgraph_api.cli:

        Welcome to

╦  ┌─┐┌┐┌┌─┐╔═╗┬─┐┌─┐┌─┐┬ ┬
║  ├─┤││││ ┬║ ╦├┬┘├─┤├─┘├─┤
╩═╝┴ ┴┘└┘└─┘╚═╝┴└─┴ ┴┴  ┴ ┴

- ?? API: http://127.0.0.1:2024
- ?? Studio UI: https://smith.langchain.com/studio/?baseUrl=http://127.0.0.1:2024
- ?? API Docs: http://127.0.0.1:2024/docs

This in-memory server is designed for development and testing.
For production use, please use LangSmith Deployment.

五、安裝langchain chat ui依賴

此步驟針對(duì)本地運(yùn)行的chat ui，如果是線上版本直接跳過(guò)

進(jìn)入langchain chat ui的解壓目錄

執(zhí)行install命令

pnpm install

運(yùn)行啟動(dòng)命令

PS E:\workspace\AGI\項(xiàng)目\agent-chat-ui> pnpm dev   # 看到一下結(jié)果表明啟動(dòng)成功過(guò)
> agent-chat-ui@0.0.0 dev E:\workspace\AGI\項(xiàng)目\agent-chat-ui
> next dev

   ▲ Next.js 15.5.14
   - Local:        http://localhost:3000
   - Network:      http://192.168.0.129:3000
   - Experiments (use with caution):
     · serverActions

 ? Starting...
 ? Ready in 2.6s

六、配置參數(shù)

打開(kāi)langchain-chat-ui地址，配置參數(shù)

Deployment UIL：本地langgraph dev命令獲取到的地址，一般為2024端口

Graph ID：langgraph.json配置文件下配置的agent名稱

點(diǎn)擊Continue即可使用