Context Window Calculator: How Much Fits?
Check whether your prompt fits an AI model’s context window. Paste text or enter tokens, pick a window size, and see tokens used, space left, and how many words and pages still fit. Free.
Context window
What the whole window holds
How to use the Context Window Calculator
- Choose the model’s context window size, or enter a custom number.
- Paste your prompt or document, or type the number of tokens you have already used.
- Add tokens for your system prompt and tools, and reserve room for the model’s response.
- Check the bar and the numbers to see how much of the window you use and how much is left.
- Use the table to see what the whole window can hold in words and pages.
Frequently asked questions
What is a context window?
A context window is the maximum amount of text, measured in tokens, that a model can consider at once. It covers your prompt, any documents, the conversation so far, and the response.
How big is the context window of current Claude models?
Anthropic’s models overview lists a 1M token context window for Claude Fable 5.1, Claude Opus 5.5, Claude Sonnet 5.5 and Claude Haiku 5.5, as last checked on 2026-11-01. Always confirm on the official page, because limits change.
How many words fit in 1 million tokens?
Anthropic says 1M tokens is roughly 555k words on its current tokenizer, which was introduced with Claude Opus 4.7. Earlier models fit about 750k words in 1M tokens. This calculator lets you pick either ratio.
How accurate is the token estimate?
It is an estimate from your word count. Real token counts depend on the tokenizer, language, code and formatting. For exact numbers use the provider’s token counting tool.
Why reserve tokens for the response?
The model’s reply uses part of the context window too, so the prompt alone can fit and the full exchange still not fit. Reserving room avoids cut-off answers.
Can I use this for other models?
Yes. Pick one of the generic window sizes or enter a custom one for any model. The window sizes for non-Claude models are not listed by name here because they change often.