ChatZhipuAI
LangChain.js 支持智谱 AI 系列模型。
¥LangChain.js supports the Zhipu AI family of models.
https://open.bigmodel.cn/dev/howuse/model
设置
¥Setup
你需要注册一个智普 API 密钥,并将其设置为名为 ZHIPUAI_API_KEY
的环境变量。
¥You'll need to sign up for an Zhipu API key and set it as an environment variable named ZHIPUAI_API_KEY
你还需要安装以下依赖:
¥You'll also need to install the following dependencies:
tip
- npm
- Yarn
- pnpm
npm install @langchain/community @langchain/core jsonwebtoken
yarn add @langchain/community @langchain/core jsonwebtoken
pnpm add @langchain/community @langchain/core jsonwebtoken
用法
¥Usage
以下是一个例子:
¥Here's an example:
import { ChatZhipuAI } from "@langchain/community/chat_models/zhipuai";
import { HumanMessage } from "@langchain/core/messages";
// Default model is glm-3-turbo
const glm3turbo = new ChatZhipuAI({
zhipuAIApiKey: "YOUR-API-KEY", // In Node.js defaults to process.env.ZHIPUAI_API_KEY
});
// Use glm-4
const glm4 = new ChatZhipuAI({
model: "glm-4", // Available models:
temperature: 1,
zhipuAIApiKey: "YOUR-API-KEY", // In Node.js defaults to process.env.ZHIPUAI_API_KEY
});
const messages = [new HumanMessage("Hello")];
const res = await glm3turbo.invoke(messages);
/*
AIMessage {
content: "Hello! How can I help you today? Is there something you would like to talk about or ask about? I'm here to assist you with any questions you may have.",
}
*/
const res2 = await glm4.invoke(messages);
/*
AIMessage {
text: "Hello! How can I help you today? Is there something you would like to talk about or ask about? I'm here to assist you with any questions you may have.",
}
*/
API Reference:
- ChatZhipuAI from
@langchain/community/chat_models/zhipuai
- HumanMessage from
@langchain/core/messages
相关
¥Related
聊天模型 概念指南
¥Chat model conceptual guide
聊天模型 操作指南
¥Chat model how-to guides