Free tool · prices checked Oct 17, 2026
LLM API cost calculator.
How much will your AI feature cost per month? Enter tokens per request and requests per day. Every model below is priced from its provider's own pricing page, with the date we checked.
Your usage
Start from an example, then change the numbers.
| Model | $ / 1M in | $ / 1M out | Per request | Per month |
|---|
Cheapest first. Click a row to pick it. Month = 30 days. Standard prices, no batch or cache discounts.
How it works
Cost per request = input tokens × input price + output tokens × output price, with prices quoted per 1 million tokens. Per month = per request × requests per day × 30.
- Long prompts cost more on some models. Where a provider lists a higher price for big prompts (over 272K tokens on several OpenAI models, over 200K on Gemini Pro models, over 100K on Claude Haiku 5.5), the calculator switches to that price when your input is above the line.
- Standard prices only. Batch processing and cached input are cheaper on all three providers, so a real bill can come in lower. Tool calls such as web search are billed on top.
- What is a token? A piece of a word. Google's Gemini docs put it at about 4 characters per token, or 60 to 80 English words per 100 tokens (source). Each model family counts tokens a bit differently.
- Thinking models. Reasoning tokens are billed as output on Gemini and OpenAI models, so set output tokens higher for models that think before they answer.
Sources
Each price was copied from the provider's official pricing page. If a price could not be checked, the model was left out.
- OpenAI API pricing, checked Oct 17, 2026
- Anthropic API pricing, checked Oct 17, 2026
- Google API pricing, checked Oct 17, 2026
Prices change often. If one looks out of date, email matt@tensorust.com and we will fix it.
Who made this
Matt Busel runs Tensorust, a one-person studio where AI agents do the distribution, so we watch these bills closely. More on LLMs: LLM articles.