AI Engineering

Batch LLM Cost Calculator

Estimate total request cost for a batch and an optional discount.

Calculations run in your browser. The tool does not send your inputs to a server.

01Your inputs

Local workspace

Start with the sample below, or enter your own values.

Input tokensEnter a numeric value using the unit shown in the label.

Output tokensEnter a numeric value using the unit shown in the label.

Input price per million tokensUse a price per 1,000,000 tokens, with the same currency for input and output.

Output price per million tokensUse a price per 1,000,000 tokens, with the same currency for input and output.

RequestsEnter a whole-number count. Check the unit in the label.

Discount (%)Enter a percentage as a number, such as 85 for 85%.

02Your result

Your result will appear here. Choose an action after checking your inputs.

A quick guide

How to use this tool

Check the example and the unit beside each input.
Enter your values and choose Calculate.
Review the result and assumptions, then copy what you need.

Reset example restores the starting values. Clear removes text and number inputs, returns choices to their first valid option, and removes the result.

Reference guide

Explanation and worked examples

Token counts are per request. The model multiplies by request count and applies the same supplied discount to input and output.

Use in day-to-day IT work

Estimate the resources for an AI-assisted IT workflow or draft its review and fallback steps.

Checks before applying the result

Use current provider limits, prices and tokenizer measurements for the intended model.
Evaluate representative support tasks, access controls and failure handling before connecting operational systems.

New to the terminology?

LLM: Large language model: a model that generates or interprets language.
Token: A unit of text counted by a model tokenizer.
RAG: Retrieval-augmented generation: supplying retrieved evidence to a model.

Keep exploring

How calculations work and where the models stop →