OpenAI API Quickstart

Use one small Responses request to verify the intended server-side integration before adding tools, streaming or application state.

#Prepare the environment

Install the official SDK and configure OPENAI_API_KEY securely. Set OPENAI_MODEL to a model ID available to the intended project; do not guess one from a marketing name.

bashCopy
pip install openai

#Python

pythonCopy
import os
from openai import OpenAI

client = OpenAI(timeout=20.0)
response = client.responses.create(
    model=os.environ["OPENAI_MODEL"],
    input="Explain a Python generator in two sentences.",
)
print(response.output_text)

#JavaScript

javascriptCopy
import OpenAI from "openai";

const client = new OpenAI({ timeout: 20000 });
const response = await client.responses.create({
  model: process.env.OPENAI_MODEL,
  input: "Explain a Python generator in two sentences."
});
console.log(response.output_text);

These are provider examples that require account authorization and may incur usage charges; they are not offline fixtures.

#Verify the minimum path

Confirm the correct project/key, a successful response and usable output. Record request identifiers and safe diagnostic metadata, while excluding secrets and sensitive prompt content from ordinary logs.

Do not begin with a complex agent or put the key in frontend code. Once the minimal request is understood, add the next capability with its own acceptance checks.

#Reference

• Responses API↗

#Related pages

• Authentication
• Chat basics
• Streaming
• OpenAI API overview