How to beginner · 3 min read

How to use stop sequences in OpenAI API

Q: How to use stop sequences in OpenAI API

Use the stop parameter in the OpenAI API's chat.completions.create method to specify one or more sequences where the model should stop generating further text. Pass a string or list of strings to stop to control output termination precisely.

Quick answer

Use the stop parameter in the OpenAI API's chat.completions.create method to specify one or more sequences where the model should stop generating further text. Pass a string or list of strings to stop to control output termination precisely.

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 for secure authentication.

bash

pip install openai>=1.0

Step by step

This example demonstrates how to use the stop parameter to halt generation when the model outputs a specific sequence, such as a newline or a custom token.

python

import os
from openai import OpenAI

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

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Write a short poem about spring."}],
    stop=["\n\n", "STOP"]
)

print(response.choices[0].message.content)

output

A gentle breeze whispers,
Flowers bloom in bright array,
Spring dances softly.

Common variations

Use a single string or a list of strings for stop to specify multiple stop sequences.
Apply stop with different models like gpt-4o-mini or gpt-4o.
In async code, pass stop the same way when calling await client.chat.completions.acreate().

python

import asyncio
import os
from openai import OpenAI

async def main():
    client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])
    response = await client.chat.completions.acreate(
        model="gpt-4o-mini",
        messages=[{"role": "user", "content": "List three fruits."}],
        stop=["."]
    )
    print(response.choices[0].message.content)

asyncio.run(main())

output

Apple
Banana
Cherry

Troubleshooting

If the model ignores your stop sequences, ensure they exactly match the output tokens including whitespace and casing.
Stop sequences can truncate output abruptly; test different sequences to find the best fit.
Using too many or very short stop sequences may cause premature stopping.

✅

Key Takeaways

Use the stop parameter to precisely control where the model output ends.
Pass a string or list of strings to stop for single or multiple stop sequences.
Stop sequences must exactly match the generated text to work correctly.
The stop parameter works consistently across synchronous and asynchronous calls.
Test and adjust stop sequences to avoid cutting off output prematurely.

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

Verify ↗