What it is
The OpenAI API gives developers programmatic access to OpenAI's models, from the GPT-6 family down to small, cheap models, plus image, audio and realtime voice models. The main interface is the Responses API, with built-in tools such as web search, file search and hosted code containers, and there are SDKs, a CLI, evals and fine-tuning. You manage keys and billing in OpenAI's developer dashboard.
It suits anyone building a product on GPT models. Billing is per million input and output tokens at each model's rate, with cheaper Batch and Flex tiers and discounts for cached input. Prices vary widely between models, and API usage is billed separately from ChatGPT subscriptions, so a ChatGPT plan does not cover it.
Best for
- Adding GPT models to your own app or service
- Building agents with hosted tools like web search and file search
- Batch jobs where a cheaper, slower tier is fine