Tokens and context
Context window fit checker
Whether a document fits a model's context window, how much of it the document uses, and how many chunks it needs otherwise.
Prices read from the vendor pages on 2026-10-20. Arithmetic on list prices, not advice or a quote.
Saving and exports need Norwell Tools Pro, $9 a month. See what is in it. Everything above is free.
How it works
A context window is the most tokens a single request may hold, prompt and answer together. Fitting is not the same as affordable: a 900K-token prompt on a 1M window is billed for 900K input tokens every call, which is where prompt caching earns its keep.
The formula
usable = window - reserve; fits = tokens <= usable; chunks = ceil(tokens / (usable * (1 - overlap)))Worked example
With the inputs the page opens on:
- document tokens: 250000 tokens
- model: claude-sonnet-5-5
- reserve tokens: 8000 tokens
- chunk overlap pct: 10%
gives
- fits: true
- context window: 1,000,000 tokens
- share of window pct: 25 %
- chunks needed: 1
- tokens per chunk: 250,000 tokens
Sources
- https://developers.openai.com/api/docs/models (read 2026-10-20)
- https://platform.claude.com/docs/en/models/overview (read 2026-10-20)
- https://ai.google.dev/gemini-api/docs/models (read 2026-10-20)
Prices read from each vendor's own pricing page on 2026-10-20; the page may have changed since. This is arithmetic on published list prices, not advice or a quote. Check the vendor's page before committing a budget.