LLM API Cost Calculator

Estimate what your AI workload costs per request and per month, and see the cheapest model first. Nothing leaves your browser.

Paste a prompt (optional – estimates input tokens)

Input tokens / request Output tokens / request Requests / day Cache hit rate % Batch API (−50%)

Model	$/1M in	$/1M out	Per request	Monthly

Standard-tier list prices, checked against official pages: . Batch = 50% off (all four providers publish half-price batch). Token estimate ≈ 4 chars/token for English.

Ad slot

How LLM API pricing works

Providers bill per million tokens, with input (your prompt and context) and output (the model's reply) priced separately. Output usually costs 3–5× more. Cost = input ÷ 1M × input price + output ÷ 1M × output price.

Prompt caching and batch discounts

Repeated prompt prefixes served from cache are billed at a fraction of the input price, often about one tenth. Batch APIs typically cost half price in exchange for slower, asynchronous results.

FAQ

Is my text uploaded? No. Everything is computed locally.
Are reasoning tokens included? Reasoning tokens are billed as output, so add them to output tokens.