
Models read text in tokens and can only consider so many at once. Send what matters, not everything, and start fresh when a chat gets long.
What a token is
Models do not read words; they read tokens — chunks of text. In English, a rough rule of thumb is that one token is about four characters, or roughly three-quarters of a word. Other languages and code can use more tokens per word.
What a context window is
The context window is how many tokens a model can consider at once: your instructions, the conversation so far, files and its own reply. Modern models have very large windows, but bigger is not automatically better.
Why it matters to you
- Cost: API pricing is usually per million tokens, for input and output separately.
- Speed: more tokens take longer to process.
- Quality: important details can get less attention when buried in very long context.

AI Beginner Bundle
Five beginner guides that take you from AI-curious to AI-confident.
$0+ · Look insidePractical rules
- Paste the relevant section, not the whole document.
- Put key instructions at the start and repeat critical ones at the end of long prompts.
- Summarize and restart long chats.
- Ask for concise output when you do not need length.
Try it
Paste a document into our Token & Cost Estimator to see roughly how many tokens it uses and how much of different context window sizes it fills.
Pricing changes often — always check the provider’s current pricing page before estimating API costs.
Put it into practice
Open the free Token & Cost Estimator: estimate tokens, context use and API cost. This article is part of The Complete Claude Guide: Models, Projects and Workflows.
Sources & further reading
Product facts were checked against official sources on 16 October 2026. AI tools change quickly, so check the provider’s current documentation for anything you rely on. Our editorial policy.


