How is the cost per request calculated?

Cost = (input tokens ÷ 1,000,000 × input price) + (output tokens ÷ 1,000,000 × output price). Providers price input and output tokens separately, and output is usually several times more expensive.

Are the prices up to date?

The model prices are editable presets based on published list prices and are clearly labelled as estimates. Providers change pricing, so confirm the current rate in your provider's dashboard before budgeting.

Is my data sent anywhere?

No. The calculator runs entirely in your browser. Nothing you enter is uploaded, stored or logged.

How do I convert words to tokens?

As a rough rule, one English word is about 1.3 tokens and 100 tokens is about 75 words. For an exact count, use a model-specific token counter.

LLM API Cost Calculator

LLM API cost calculator — per request and per month

Estimate what an OpenAI, Anthropic or Google AI API call will actually cost before you ship. Pick a model, enter the input and output tokens for a typical request, add your monthly volume, and see the cost per call and the projected monthly bill update instantly.

How LLM pricing works

Large language models bill per token, and almost always price input (your prompt) and output (the model’s completion) at different rates — output is typically 3–5× more expensive. Prices are quoted per million tokens, so the cost of one request is:

cost = (input_tokens / 1,000,000) × input_price
     + (output_tokens / 1,000,000) × output_price

A token is roughly ¾ of an English word, so 1,000 tokens ≈ 750 words.

Tips to cut your bill

Trim the prompt. Long system prompts are billed on every call — move stable context into a cached or shorter form.
Cap output. Set max_tokens so a runaway completion can’t 10× your cost.
Right-size the model. A mini/flash model is often 10–20× cheaper and good enough for classification, extraction and routing.
Batch and cache. Prompt caching and batch APIs can cut input costs sharply at volume.