Token Counter

Paste any text and this token counter gives you an estimated token count alongside words and characters, then shows what share of common context windows it would fill, from 8K up to 1M tokens. The estimate is based on character and word counts, not a specific provider's tokenizer, so treat it as a working estimate rather than an exact figure.

Tokens are the units language models actually read, and every model has a limit on how many it can hold at once. Knowing roughly where a document lands tells you whether it will fit in one message, how much room is left for the answer, and what an API call might cost. Everything is computed in your browser: your text is never uploaded.

How to use it

1. Paste or type your text into the box.
2. Read the estimated token count, along with words and characters, as you edit.
3. Check the context window table to see how much of each common limit the text would use.

What is a token?

Language models do not read letters or whole words. They read tokens: chunks of text produced by a tokenizer that turns common words into a single token and rarer words, numbers, and punctuation into several. "The" is one token; an unusual surname or a long URL might be five or more. Each provider trains its own tokenizer, so the same sentence can cost a slightly different number of tokens on GPT, Claude, and Gemini.

Tokens matter for two reasons. Context windows are measured in tokens, so they decide how much a model can read and remember in one conversation. API pricing is also per token, so the count drives cost. Because this tool estimates from characters and words rather than running a provider's tokenizer, the number is a planning figure, not the exact count you would be billed for.

How many tokens is a page of text?

For ordinary English prose, a token works out to roughly three quarters of a word, or about four characters. Code, other languages, and text full of numbers or symbols usually cost more tokens per word. These rough conversions are enough for planning:

Leave room for the answer. If a document fills most of a context window, the model has little space left to reply, and long conversations shrink that space further with every turn.

• 1 token ≈ 4 characters, or about 0.75 words of English
• 100 tokens ≈ 75 words, a short paragraph
• 1,000 tokens ≈ 750 words, roughly a page and a half single-spaced
• 8K tokens ≈ 6,000 words, a long article or a short story
• 32K tokens ≈ 24,000 words, a novella or a lengthy report
• 128K tokens ≈ 96,000 words, a full-length novel
• 200K tokens ≈ 150,000 words, a long novel or a thick contract
• 1M tokens ≈ 750,000 words, a shelf of books or a large codebase