Upstash Redis-Backed Chat Memory

Because Upstash Redis works via a REST API, you can use this with Vercel Edge, Cloudflare Workers and other Serverless environments. Based on Redis-Backed Chat Memory.

For longer-term persistence across chat sessions, you can swap out the default in-memory chatHistory that backs chat memory classes like BufferMemory for an Upstash Redis instance.

Setup

You will need to install @upstash/redis in your project:

See this section for general instructions on installing integration packages.

npm

npm install @langchain/openai @langchain/community @langchain/core @upstash/redis

You will also need an Upstash Account and a Redis database to connect to. See instructions on Upstash Docs on how to create a HTTP client.

Usage

Each chat history session stored in Redis must have a unique id. You can provide an optional sessionTTL to make sessions expire after a give number of seconds. The config parameter is passed directly into the new Redis() constructor of @upstash/redis, and takes all the same arguments.

import { BufferMemory } from "langchain/memory";
import { UpstashRedisChatMessageHistory } from "@langchain/community/stores/message/upstash_redis";
import { ChatOpenAI } from "@langchain/openai";
import { ConversationChain } from "langchain/chains";

const memory = new BufferMemory({
  chatHistory: new UpstashRedisChatMessageHistory({
    sessionId: new Date().toISOString(), // Or some other unique identifier for the conversation
    sessionTTL: 300, // 5 minutes, omit this parameter to make sessions never expire
    config: {
      url: "https://ADD_YOURS_HERE.upstash.io", // Override with your own instance's URL
      token: "********", // Override with your own instance's token
    },
  }),
});

const model = new ChatOpenAI({
  model: "gpt-3.5-turbo",
  temperature: 0,
});

const chain = new ConversationChain({ llm: model, memory });

const res1 = await chain.invoke({ input: "Hi! I'm Jim." });
console.log({ res1 });
/*
{
  res1: {
    text: "Hello Jim! It's nice to meet you. My name is AI. How may I assist you today?"
  }
}
*/

const res2 = await chain.invoke({ input: "What did I just say my name was?" });
console.log({ res2 });

/*
{
  res1: {
    text: "You said your name was Jim."
  }
}
*/

Advanced Usage

You can also directly pass in a previously created @upstash/redis client instance:

import { Redis } from "@upstash/redis";
import { BufferMemory } from "langchain/memory";
import { UpstashRedisChatMessageHistory } from "@langchain/community/stores/message/upstash_redis";
import { ChatOpenAI } from "@langchain/openai";
import { ConversationChain } from "langchain/chains";

// Create your own Redis client
const client = new Redis({
  url: "https://ADD_YOURS_HERE.upstash.io",
  token: "********",
});

const memory = new BufferMemory({
  chatHistory: new UpstashRedisChatMessageHistory({
    sessionId: new Date().toISOString(),
    sessionTTL: 300,
    client, // You can reuse your existing Redis client
  }),
});

const model = new ChatOpenAI({
  model: "gpt-3.5-turbo",
  temperature: 0,
});

const chain = new ConversationChain({ llm: model, memory });

const res1 = await chain.invoke({ input: "Hi! I'm Jim." });
console.log({ res1 });
/*
{
  res1: {
    text: "Hello Jim! It's nice to meet you. My name is AI. How may I assist you today?"
  }
}
*/

const res2 = await chain.invoke({ input: "What did I just say my name was?" });
console.log({ res2 });

/*
{
  res1: {
    text: "You said your name was Jim."
  }
}
*/