Skip to content
MyStackPlan
LLM APIs & Inference

OpenAI API pricing

The GPT models behind ChatGPT, available by API and billed per token.

Try it in the builderOfficial pricing pageReport a wrong price

Every plan below was checked against OpenAI API's own pages on . Prices in USD.

Plans

Pay as you go

Pay as you go

At the limit
Bills for extra usage
  • Billed per million tokens, with input and output priced separately. Prices shown are standard short-context rates.
  • Cached input is much cheaper, and long-context requests cost more.
  • Regional processing (data residency) endpoints add 10% for models released on or after March 5, 2026.

Source: developers.openai.com · checked

Models

ModelInput per 1M tokensOutput per 1M tokensAnnounced change
GPT-6 LunaCheapest current model$0.10$0.50None announced
GPT-6.1 Sol$2.00$10.00None announced
GPT-6 AstraMost capable, most expensive$10.00$50.00None announced

Works well with

  • Cloudflare AI GatewayCloudflare AI Gateway can sit in front of OpenAI requests to add logging, caching and rate limits.
  • OpenRouterOpenRouter lets you call OpenAI models through one API.
  • Vercel AI GatewayVercel AI Gateway lets you call OpenAI models through one gateway.

Alternatives for llm apis & inference