How many words is 1 million tokens?

One million tokens generally translates to roughly 700,000 to 750,000 words in English, based on the commonly used estimate that one token equals about 0.75 words. This ratio comes from how large language models break text into tokens, which are often smaller than full words, like word fragments, common prefixes, suffixes, or even individual punctuation marks. So while it's tempting to think of tokens as a one-to-one match with words, the actual count depends heavily on the type of content you're processing. Simple, everyday English text with short common words will lean toward the higher end of that word estimate, while technical writing, code, or text full of unusual terms and non-English characters will use more tokens per word, pushing the actual word count lower. To put a million tokens into perspective, that's roughly equivalent to about 1,500 typical pages of double-spaced text, or several full-length novels combined. This is why context window sizes are usually described in tokens rather than words, since it's a more consistent unit of measurement across different languages and types of content being processed.