How to beginner · 3 min read

How to implement conversation history

Quick answer

Implement conversation history by storing chat messages as a list of messages with roles like user and assistant. Pass this list to the client.chat.completions.create method to maintain context across turns.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install openai>=1.0

Setup

Install the official OpenAI Python SDK and set your API key as an environment variable.

bash

pip install openai>=1.0

output

Collecting openai
  Downloading openai-1.x.x-py3-none-any.whl (xx kB)
Installing collected packages: openai
Successfully installed openai-1.x.x

Step by step

Store conversation history as a list of messages and pass it to the chat completion call to maintain context.

python

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

# Initialize conversation history
conversation_history = [
    {"role": "system", "content": "You are a helpful assistant."}
]

# User sends a message
user_message = "Hello, who won the world series in 2020?"
conversation_history.append({"role": "user", "content": user_message})

# Get assistant response
response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=conversation_history
)
assistant_message = response.choices[0].message.content
conversation_history.append({"role": "assistant", "content": assistant_message})

print("Assistant:", assistant_message)

# Next user message
next_user_message = "Where was it played?"
conversation_history.append({"role": "user", "content": next_user_message})

# Continue conversation with full history
response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=conversation_history
)
assistant_message = response.choices[0].message.content
conversation_history.append({"role": "assistant", "content": assistant_message})

print("Assistant:", assistant_message)

output

Assistant: The Los Angeles Dodgers won the World Series in 2020.
Assistant: The 2020 World Series was played at Globe Life Field in Arlington, Texas.

Common variations

You can implement conversation history with other models like gpt-4o-mini or use async calls. For streaming, pass stream=True and process chunks incrementally.

python

import asyncio
import os
from openai import OpenAI

async def async_chat():
    client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])
    conversation_history = [
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Hello!"}
    ]

    # Async streaming response
    stream = await client.chat.completions.create(
        model="gpt-4o-mini",
        messages=conversation_history,
        stream=True
    )

    print("Assistant:", end=" ")
    async for chunk in stream:
        delta = chunk.choices[0].delta.content or ""
        print(delta, end="", flush=True)
    print()

asyncio.run(async_chat())

output

Assistant: Hello! How can I assist you today?

Troubleshooting

If the assistant forgets context, ensure you pass the full conversation_history list each time.
Trim history if it grows too large to avoid token limits.
Use roles correctly: system, user, and assistant.

✅

Key Takeaways

Maintain conversation history as a list of messages with roles to preserve context.
Always pass the full message history to client.chat.completions.create for coherent multi-turn dialogs.
Use streaming and async calls for responsive chat applications.
Manage token limits by trimming or summarizing long histories.
Roles system, user, and assistant are essential for proper context handling.

Verified 2026-04 · gpt-4o-mini

Verify ↗