TokenTally is a free, independent calculator for comparing the cost of large-language-model APIs. It exists because provider pricing pages are scattered, formatted differently, and rarely answer the question developers actually have: "what will my workload cost per month, on each model?"
We track per-token prices for models from OpenAI, Anthropic, Google, xAI, DeepSeek, Mistral and Meta, and model the two discounts that matter in practice — prompt caching and batch processing. Prices are compiled from providers' public pricing pages and refreshed regularly; the last-verified date is always shown in the site header.
TokenTally is not affiliated with, sponsored by, or endorsed by any AI provider. Nothing here is financial advice — always confirm current prices on the provider's official page before making commitments.
Questions, corrections, or a price change we missed? Email shduswo0530+tokentally@gmail.com.