OpenAI

OpenAI's API for GPT models via the Responses and Chat Completions APIs. Supports chat, reasoning, vision, tool calling, streaming, speech-to-text, realtime voice, image generation and embeddings; official SDKs for Python and JavaScript/TypeScript.

Website Docs Changelog Compare with others

OpenAI at a glance

From official pages, checked Sep 29, 2026. Tap the arrow for each source.

Starting price: $2 per 1M input tokens
GPT-6-Sol; $10 per 1M output tokens (GPT-6-Luna from $0.10 input).
Rate limit: 500 requests/min
Tier 1 ($5 paid), GPT-6-Sol; 500,000 tokens/min.
Size limit: 1.05M token context
GPT-6 Astra, Sol and Luna.
Storage: 1 GB free
File search storage, then $0.10 per GB per day.
Official SDKs: PythonJavaScript / TypeScript
Trains on your data: No
API data is not used for training unless you explicitly opt in.
EU hosting: Yes
Europe (EEA + Switzerland) data residency via eu.api.openai.com; requires approval.
Compliance: SOC 2ISO 27001GDPRHIPAA

What you can build with OpenAI

OpenAIRecipeBuild

Stream GPT-6 Astra responses from a Next.js route handler

Pipe OpenAI Responses API text deltas through a Next.js App Router endpoint so replies render token by token.

From $2 per 1M input tokens

Chat

OpenAI at a glance

What you can build with OpenAI

Stream GPT-6 Astra responses from a Next.js route handler

Update timeline