A token is a small piece of text, roughly a word or part of one, that an AI model reads and writes in. Tokens are also the unit an agent's working memory is measured in, so the size of its context window is given as a number of tokens.
Token
How tokens work
AI models don't read text as whole sentences; they break it into tokens, small chunks roughly the size of a word or a fragment of one, and process those. Everything an agent holds in memory, your conversation, the code so far, and the decisions made, takes up a number of tokens, and each agent has a fixed budget called its context window. On Emergent, agents work with a large context window of about 200,000 tokens, enough for most full-stack builds, and when a long project starts to fill it, Forking lets you continue in a fresh window while keeping your code.
Examples
A short prompt uses only a handful of tokens, while a long build that has accumulated lots of conversation and code can use most of a 200,000-token context window. When an Emergent project approaches that token limit, you fork it to keep building with room to spare, rather than running out of memory mid-project.
Frequently Asked Questions
Your Questions, Answered
This will automatically populate, don't change
Don't change this element unless you know what you are doing
Start Building
on Emergent today
on Emergent today
Start Building
