How to beginner · 3 min read

How to use ConversationChain in LangChain

Q: How to use ConversationChain in LangChain

Use ConversationChain from langchain.chains to build conversational AI flows by combining an LLM and a memory component. Initialize it with a chat model like ChatOpenAI and a memory such as ConversationBufferMemory to maintain context across turns.

Quick answer

Use ConversationChain from langchain.chains to build conversational AI flows by combining an LLM and a memory component. Initialize it with a chat model like ChatOpenAI and a memory such as ConversationBufferMemory to maintain context across turns.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install langchain_openai>=0.2.0

Setup

Install the required packages and set your OpenAI API key as an environment variable.

bash

pip install langchain_openai

Step by step

This example shows how to create a ConversationChain with ChatOpenAI and ConversationBufferMemory to maintain chat history.

python

import os
from langchain_openai import ChatOpenAI
from langchain.chains import ConversationChain
from langchain.memory import ConversationBufferMemory

# Initialize chat model
chat_model = ChatOpenAI(model="gpt-4o-mini", temperature=0.7)

# Initialize memory to store conversation history
memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True)

# Create ConversationChain with model and memory
conversation = ConversationChain(llm=chat_model, memory=memory)

# Run conversation with user input
response1 = conversation.invoke({"input": "Hello, who won the World Series in 2023?"})
print("Bot:", response1)

# Continue conversation, memory keeps context
response2 = conversation.invoke({"input": "Where was it played?"})
print("Bot:", response2)

output

Bot: The Texas Rangers won the 2023 World Series.
Bot: The 2023 World Series was played at Globe Life Field in Arlington, Texas.

Common variations

Use different chat models like gpt-4o or gpt-4o-mini for cost/performance tradeoffs.
Use ConversationSummaryMemory for long conversations to summarize context.
Run asynchronously with await conversation.ainvoke({"input": "..."}) if using async models.

Troubleshooting

If you get authentication errors, ensure OPENAI_API_KEY is set correctly in your environment.
If context is not remembered, verify you passed memory to ConversationChain.
For unexpected responses, adjust temperature or switch to a more capable model.

✅

Key Takeaways

Use ConversationChain with ChatOpenAI and memory to maintain chat context.
Choose memory types like ConversationBufferMemory or ConversationSummaryMemory based on conversation length.
Adjust model and temperature parameters for desired response style and cost.
Always set your API key via environment variables for security.
Async invocation is available for async-compatible chat models.

Verified 2026-04 · gpt-4o-mini, gpt-4o

Verify ↗