LLM API Cost Calculator

Estimate what an AI feature costs before you ship it, see how much of the bill is output tokens, and compare two models side by side. Computed live in your browser.

Your workload

Prices are examples only. Enter the current per-million-token prices from each provider's pricing page, since they change over time.

Model A
—
per month
Per request
—
Per year
—
Input share
—
Output share
—
Model B
—
per month
Per request
—
Per year
—
Input share
—
Output share
—
—

How to use this

Output tokens usually cost several times more than input tokens, so the input and output split often surprises people: a chatty feature can cost far more than a terse one at the same volume. Estimate tokens as roughly three quarters of a word each, or paste a real prompt into a provider's tokenizer for a precise count. The two-model comparison is the quickest way to see whether a cheaper tier would do the job: run a frontier model in column A and a fast tier in column B.

This is how you catch a feature that is fine in a demo but ruinous at scale. To pick the right tier and version, use the AI model selector. Ways to cut the bill without hurting quality are in LLM cost optimization, and running a cheap model first with escalation is in multi-model architecture.

Built by Deepanshu Grover, a Growth Product Manager and AI builder in Paris. Runs entirely in your browser, nothing is sent anywhere. About · Get in touch