Docs

ReadHelp

On this page

Context window

Your AI can keep only so much of a chat in mind at once. That room is its context window. It is counted in tokens, which are pieces of words.

The ring beside the chat

The ring next to where you type shows how full this chat's context window is. Point at it or tap it to see the numbers: how many tokens the chat holds, and how many fit.

When it is full

Your next message starts a fresh session. Before it does, your AI writes itself a short hand-off note: what you want, what was decided, what is still open, and anything it promised to do. The fresh session reads that note and your latest messages, then carries on. You keep seeing one chat.

A chat also starts fresh after it has been quiet for a while, the same way.

What it remembers

A fresh session does not start from nothing. As you work, your AI saves what matters to your memory: decisions, facts about your business, and what got done. Every session reads your memory first.

A small detail said once in a long chat can still slip. If your AI seems to have forgotten something, say it again, or ask it to save it to memory.

Not your plan's limit

The context window is how much of one chat your AI holds at once. Your AI plan has its own limit on how much your AI can do in a given time. The two fill up separately.