Context Window Calculator
Enter the length of a document in characters or words. It is converted to tokens and checked against the context window you pick, leaving room for the answer. Token counts are estimates, so the math uses a range.
Check your model's window size in the official docs.
Output tokens share the same window. Leave as much as the answer you expect.
Fits with room to spare
Fits with room to spare
- Document tokens (estimate)
- 8,000 ~ 23,000 tokens
- Room for input
- 124,000 tokens
- Share of the window (high estimate)
- 18.0%
- How many such documents fit
- 5
English uses Google's official rule of thumb: a token is about 4 characters, and 100 tokens are about 60 to 80 English words. There is no official ratio for Korean, so a character (spaces included) is taken as 0.4 to 1.15 tokens. The window buttons are common sizes, not the values of any specific model.
Frequently asked questions
What is a context window?
The total number of tokens a model can read and write in one go. The prompt, attached documents, earlier messages, and the model's answer all have to fit in it.
Why leave room for the answer?
Output tokens share the same window. If the document fills it, there is no room left to answer, and the reply can be cut off or the request rejected. Models also have a separate maximum output, so check that in the official docs too.
Which model do the window buttons belong to?
None in particular. They are sizes you often see. Window sizes differ by model and plan and change often, so look up your model in the official docs and enter it.
If a long document fits, will the model use all of it well?
Fitting and using it well can differ. With very long inputs, the middle may get less attention and cost and latency go up, so sending only the parts you need is often better.
Researchlog collects official AI announcements as summary cards
Only the operator uses it for now, so the updates screen stays closed even after sign-up. You can see what it does first.