LLM cost calculator
Estimate your monthly API bill across 321+ models. Set your tokens and volume — or pick a preset — and we'll rank every model by what it would cost you.
That's 75M tokens/month total (50M in + 25M out).
Estimates use list API prices and ignore cached-input discounts, batching and free tiers. Showing the 40 cheapest of 321 priced models for your settings.
How LLM pricing works
How do I calculate LLM API costs?
API cost = (input tokens × input price + output tokens × output price) ÷ 1,000,000, multiplied by your number of requests. This calculator does it for every model at once so you can compare. A token is roughly ¾ of a word.
What is a blended price?
A single number that mixes input and output price at a 3:1 ratio (the typical chat workload). It's handy for ranking, but for an accurate estimate use your real input:output split above — output tokens usually cost 3-5× more than input.
Are these the exact prices I'll pay?
They're list API prices from OpenRouter, updated hourly. Your actual bill can be lower with cached-input discounts, batch pricing or free tiers, and higher with image or tool-call overhead. Treat these as a close estimate.
See the full pricing table · cheapest models · leaderboard