Cost calculator

Estimate what a workload costs on Claude and ChatGPT models, and how much caching and batching save.

Input tokens per request Output tokens per request Requests per month

Input served from cache: 0% Batch API (50% off)

Model	Per request	Per month	Without savings

API list prices in USD — check official pricing for Claude and OpenAI before relying on them. Cached input is billed at the cache-read rate; one-time cache write premiums are not included. The ChatGPT app itself is a subscription and is not billed per token. The same text uses a slightly different number of tokens in each model; measure yours with the token counter.