OpenAI API Platform

What it does

The OpenAI API Platform takes developer requests containing text, images, audio, or tool definitions and returns model-generated output, such as text, structured JSON, images, transcriptions, or real-time audio responses. Developers call it over HTTP or through official SDKs in JavaScript, Python, Go, Java, C#, and Ruby, and a command-line interface is also shown in the documentation.

Developers use the platform to build their own applications rather than to chat in a finished app. The Responses API handles direct model requests with tools and multimodal input, while the Agents SDK adds code-first agents with tool use, handoffs, approvals, and tracing. Several models are offered at different tiers of capability, speed, and cost, and built-in tools such as web search and file search are billed on top of model tokens.

The platform suits developers and technical teams who can write code and manage API keys. Access requires an OpenAI account and an API key, and usage is metered per token, per call, or per minute depending on the model or tool. Costs can rise quickly with high volume, and the documentation notes that some features, such as certain fine-tuning and specialized models, have restricted or wound-down access.

What it can do

• Generate text and structured JSON output from a prompt via the Responses API.
• Analyze images as input to models with vision capabilities.
• Generate images with GPT Image models.
• Transcribe and generate audio, including real-time audio sessions.
• Build tool-using agents with the Agents SDK, including handoffs, approvals, and tracing.
• Use built-in tools such as web search, file search, and hosted code execution containers.

Works with

• Amazon Bedrock (OpenAI models billed through Amazon Bedrock)
• Microsoft Azure (OpenAI models billed through Microsoft Azure)

Example workflows

Generate a short story from a prompt

Input: a text prompt such as a request for a bedtime story. Steps: install an official SDK, set OPENAI_API_KEY, call the Responses API with a model name and the input text. Output: generated text read from the response's output text field.

Build a tool-using agent

Input: a task that needs external tools or handoffs between agents. Steps: define agents and tools with the Agents SDK, run the agent with an API key, and review tracing for approvals and tool calls. Output: a final response plus a trace of tool and handoff steps.

Before you choose

• Pricing is per token, per call, or per minute depending on the model or tool; check the pricing page for the specific model and feature before estimating costs.
• Some features are restricted or being wound down: fine-tuning is no longer open to new users, and certain life-sciences models are limited to approved internal research access.
• Cloud-hosted OpenAI models on Amazon Bedrock and Microsoft Azure are billed through those services rather than OpenAI.

Listed under

• Coding
• AI models
• API
• App development
• Developer API
• Developers
• Language models

Get started

The documentation offers a developer quickstart and an option to create an API key.

1. Create an OpenAI account and generate an API key from the platform console.
2. Export the key as an environment variable such as OPENAI_API_KEY, or pass it to the SDK client.
3. Install an official SDK for your language or use the CLI, then send a first request to the Responses API.

Billed by usage with no subscription plan shown on the page. Prices are per 1M tokens for language models, per minute for transcription and voice sessions, and per 1,000 calls for web search. Batch, Flex, Fast, and Ultrafast tiers have different rates. Regional processing endpoints carry a 10% uplift for data-residency-eligible models released on or after March 5, 2026. FedRAMP endpoints also carry a 10% uplift starting November 1, 2026. Free tier and trial terms are not confirmed in the reviewed information.

Source: developers.openai.com/api/docs/pricing, checked 1 Nov 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 1 Nov 2026. Each source supports part of the overview; availability and prices can change.

• OpenAI API Platform Documentation ↗developers.openai.com
• Pricing | OpenAI API ↗developers.openai.com

Company website

openai.com: product, pricing and policy pages

Web archive

When the site first appeared

YC directory mirror

Batch and company name

Web search

Confirms the official site

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

Last checked 1 Nov 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.