How to beginner · 3 min read

How to extract key points from text with AI

Q: How to extract key points from text with AI

Use the OpenAI Python SDK to send your text to a chat model like gpt-4o with a prompt instructing it to extract key points. Parse the response from client.chat.completions.create to get concise bullet points or summaries.

Quick answer

Use the OpenAI Python SDK to send your text to a chat model like gpt-4o with a prompt instructing it to extract key points. Parse the response from client.chat.completions.create to get concise bullet points or summaries.

PREREQUISITES

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

Setup

Install the official openai Python package and set your API key as an environment variable.

Run pip install openai to install the SDK.
Set your API key in your shell: export OPENAI_API_KEY='your_api_key' (Linux/macOS) or setx OPENAI_API_KEY "your_api_key" (Windows).

bash

pip install openai

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

This example sends a text passage to the gpt-4o chat model with a prompt to extract key points. It prints the bullet points returned by the model.

python

import os
from openai import OpenAI

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

text_to_summarize = (
    "Artificial intelligence (AI) is transforming industries by automating tasks, "
    "enhancing decision-making, and enabling new capabilities. Key benefits include increased efficiency, "
    "cost reduction, and improved accuracy. Challenges remain in ethics, bias, and data privacy."
)

prompt = (
    f"Extract the key points from the following text as concise bullet points:\n\n{text_to_summarize}"
)

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": prompt}]
)

key_points = response.choices[0].message.content
print("Key points extracted:\n", key_points)

output

Key points extracted:
- AI automates tasks and transforms industries.
- Benefits include increased efficiency, cost reduction, and improved accuracy.
- Challenges include ethics, bias, and data privacy.

Common variations

Use gpt-4o-mini for faster, cheaper extraction with slightly less detail.
Use async calls with asyncio and await client.chat.completions.create(...) for concurrency.
Stream partial results by setting stream=True and iterating over chunks.
Use Anthropic's claude-3-5-sonnet-20241022 model with the anthropic SDK for alternative summarization.

python

import asyncio
import os
from openai import OpenAI

async def extract_key_points_async(text: str):
    client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])
    prompt = f"Extract the key points from the following text as concise bullet points:\n\n{text}"
    response = await client.chat.completions.create(
        model="gpt-4o-mini",
        messages=[{"role": "user", "content": prompt}],
        stream=True
    )
    async for chunk in response:
        print(chunk.choices[0].delta.content or "", end="", flush=True)

if __name__ == "__main__":
    text = "AI improves efficiency, reduces costs, and raises ethical concerns."
    asyncio.run(extract_key_points_async(text))

output

- AI improves efficiency
- reduces costs
- raises ethical concerns

Troubleshooting

If you get authentication errors, verify your OPENAI_API_KEY environment variable is set correctly.
If the model returns irrelevant or verbose output, refine your prompt to be more explicit about concise bullet points.
For rate limits, implement exponential backoff retries or upgrade your API plan.
Ensure you use the latest openai SDK version to avoid deprecated method errors.

✅

Key Takeaways

Use the OpenAI Python SDK with gpt-4o to extract key points via chat completions.
Craft clear prompts requesting concise bullet points for best summarization results.
Async and streaming calls improve performance for large or multiple texts.
Always set your API key securely via environment variables to avoid authentication issues.

Verified 2026-04 · gpt-4o, gpt-4o-mini, claude-3-5-sonnet-20241022

Verify ↗