LLM API Monthly Cost Comparison
Enter your monthly usage to see what several models cost officially and at the cheapest provider, with cache and tiered prices included, plus the plans that include them.
How it is calculated
Cached input is charged at the cache read price, or the input price when none is published. Uncached input is charged at the cache write price when it is written to the cache. When the average input per request exceeds a tier threshold, the whole request uses that tier. Per-request prices are converted with monthly input ÷ average input per request. The lowest monthly cost is the cheapest row across all providers and groups, and the usage ratio = lowest monthly cost ÷ official monthly cost. How usage estimates work
Why plans are not converted
Coding Plan and subscription quotas are counted in requests, credits or relative multiples, and the tokens per request differ by provider, so they cannot be converted into token usage. This page only lists how many plans include a model and the lowest monthly-equivalent price. See the plan pages for each quota as published.
See one model in detail
Every model page has a usage estimate that uses the same usage to list the monthly cost and calculation for every provider and group. Usage you change here applies there too.