Usage limit
In shortA usage limit caps how much you can use an AI chat plan in a set period, such as a session or week, before you must wait or pay more.
In plain words
Chat subscriptions with a flat monthly price still have a ceiling. The provider limits how much you can use in a period, because each answer costs computing power. This is the usage limit. It is separate from the rate limit of an API, which caps speed rather than a plan's total use.
Anthropic's help article is a good example. Paid Claude plans have a session limit that resets every five hours and weekly limits that reset once a week. Usage in the web app, Claude Code and the desktop app counts toward the same limit. Other providers set their own periods and rules, so read your plan's page.
Why it matters
Anthropic names what makes the allowance run out faster:
- the length and complexity of your conversations,
- the features you use,
- the model you choose,
- the effort level you select.
A longer chat is more costly because each reply carries the earlier turns along. This ties to the context window and the number of tokens processed. Higher plan tiers offer larger allowances.
If you hit a limit, you can wait for the reset or upgrade your plan. The new plan's limits apply at once, as Anthropic describes it. With usage credits enabled, work continues on those credits instead of stopping. Plans and prices are compared in AI subscriptions.
If you build software, you do not use a chat plan but the API, which bills per token with its own limits. See subscription vs API. Anthropic's API also has monthly spend limits, which cap money for API use.
Example
You work on a long document for an afternoon in one chat with the largest model. By late afternoon you see a notice that you reached your session limit and when it resets. You can switch to a smaller model for light tasks, start a fresh chat for a new topic, or wait for the reset.
Common misconceptions
- "Unlimited plans are unlimited." Check the fine print. Many plans cap use or slow you down at peak load.
- "Short messages cost little, always." The cost depends on the whole conversation that is processed, not only your last line.
- "Usage limit and rate limit are the same." A rate limit controls API speed. A usage limit controls how much a plan can be used in a period.
Questions and answers
Why did my AI chat stop working?
You probably reached the usage limit of your plan for the current period. The app usually shows when the limit resets.
How can I make my usage last longer?
Start new chats for new topics, send less text, pick a smaller model for simple tasks and avoid features you do not need.
Does a usage limit reset?
Yes. Providers reset limits after a set period, for example every few hours or once a week. Check your plan for the exact rule.
Read next
- Rate limit
A rate limit caps how many requests or tokens you may send to an AI API in a set time; going over returns an error, usually HTTP 429.
- AI subscriptions compared: monthly plans and limits
An overview of AI chat subscriptions: monthly price, what is included, and usage limits.
- Subscription or API: what is cheaper for you?
A flat $20 plan or pay per token: one formula and three examples show where each wins, plus the limits and privacy settings that count beyond cost.
- Context window
The context window is the maximum amount of text, counted in tokens, that a model can read and write in one request.
- Token
A token is a small chunk of text, often a word piece, that an AI model reads, writes and bills.