Skip to main content

Basic text generation

Send a prompt to the /v1/responses endpoint:

With reasoning

Enable reasoning for complex tasks:

OpenAI-compatible format

Use /v1/chat/completions if you prefer the OpenAI message format:

Temperature guide