Models & Pricing

All active models, grouped by modality, with rates in the relevant unit for each. Pulled directly from GET /models — always in sync with the live catalog, with no hardcoded prices.

Loading live pricing…

How to read the rates

• Text — rate per 1 million tokens, separately for input and output.
• Image — rate per megapixel; cost = (width × height / 1M) × rate × number of images.
• Transcribe — gpt-4o-transcribe is billed per audio token + output text tokens.

Billing details are in Billing & Pricing.