Compare input, output, and cache-hit costs across leading providers and models — quickly find the best value, pick what you need, and call it in one click.
Model prices could not be loaded right now.
Sanki is purely pay-as-you-go. Every API call deducts from your balance based on the tokens actually consumed, with separate unit prices for input, output and cached-token reads. All rates are published per 1M tokens on this page and in the Model Square — no monthly rent, no minimum spend.
Sign up free, get your API key instantly, and call 200+ models through one API.

检测到您的浏览器使用中文,要切换到中文版吗?