Lesson 303 of 1596
The Responses API: OpenAI's Modern Developer Surface
The Responses API is where OpenAI puts stateful conversations, multimodal inputs, tools, and structured outputs. Learn the shape before you build.
Creators · Tools Literacy · ~27 min read
One Surface for Text, Vision, State, and Tools
Chat Completions taught a generation of developers the messages array. The Responses API is the newer shape: it supports text and image inputs, text outputs, conversation state, built-in tools, and function calling in one API.
The simplest Responses call still looks like a normal model request.
import OpenAI from "openai"; const client = new OpenAI(); const response = await client.responses.create({ model: "gpt-5.5", instructions: "Be concise and verify assumptions.", input: "Draft an incident-summary template for a failed deploy.", }); console.log(response.output_text);Compare the options
| Use Chat Completions when | Use Responses when |
|---|---|
| You maintain a stable legacy chat app | You need built-in tools or multimodal workflows |
| You cannot migrate everything yet | You need conversation state without rebuilding threads |
| The endpoint is already working and low risk | You are starting a new OpenAI integration |
| You only need simple message in, text out | You need structured output, tools, or future agent features |
- 1Wrap model calls behind one local client function.
- 2Move one low-risk flow to Responses.
- 3Add logging for latency, cost, and output quality.
- 4Then migrate flows that need tools, vision, or structured data.
- 5Keep old Chat Completions flows until they have a reason to move.
Key terms in this lesson
The big idea: new OpenAI apps should usually start with Responses, while old apps should migrate where the newer surface pays for itself.
End-of-lesson quiz
Check what stuck
8 questions · Score saves to your progress.
Tutor
Curious about “The Responses API: OpenAI's Modern Developer Surface”?
Ask anything about this lesson. I’ll answer using just what you’re reading — short, friendly, grounded.
Progress saved locally in this browser. Sign in to sync across devices.
Related lessons
Keep going
Creators · 45 min
OpenAI Use-Case Playbook: Match the Surface to the Job
OpenAI now spans chat, coding agents, APIs, images, realtime voice, search, files, and tools. Learn which surface belongs to which kind of product.
Creators · 11 min
AI cost attribution tools
Attribute LLM spend to teams, features, and customers.
Creators · 11 min
AI tool call debugging tools
Debug why an agent picked the wrong tool or wrong arguments.
