Context Window Calculator

Check whether a prompt plus expected completion fits a model context window. If expected output is blank, the tool assumes output ≈ 25% of input tokens and labels that assumption.

Context sizes come from the same versioned model file as the Token Counter. Last-verified date is printed in the JSON result.

Input
Paste or type your content here…
Results

Paste a prompt to see how much of the context window it would use.

Frequently Asked Questions

What is the GPT-4o context window size?
See the model dropdown and the result JSON. Figures are snapshot values with a last-verified date.
Can I reuse a token-counter paste?
The token counter stores a local handoff. You can also share via the URL data parameter.