Model cost comparison

One prompt, one expected output length — costed across every model at once. Paste your prompt to auto-count input tokens with a real tokenizer, set the output length, and sort the table to find the cheapest option for your input/output mix.

Model Input /1M Output /1M Input $ Output $ Total / call

Cheapest highlighted. Rates as of June 2026 (checked 2026-06-06), standard list prices excluding caching/batch discounts — verify before budgeting. Input counts are exact for OpenAI, estimates elsewhere (same token count is applied to every model for an apples-to-apples comparison).

More free token tools