How do Claude Code usage limits work, and how do I check my usage?

Short answer

On Pro and Max, Claude Code shares a rolling 5-hour limit and a weekly limit with Claude chat; Team and seat-based Enterprise seats work the same way per seat. With an API key there is no plan limit: you pay per token. Type /usage in Claude Code (/cost and /stats are aliases) to see plan usage bars, session cost and what eats your allowance.

Which limit applies to you

How Claude Code is metered depends on how you signed in, not on which surface you use. As of November 2026:

How you sign in What limits you Shared with
Pro A 5-hour session limit and a weekly limit Claude chat, IDE use, cloud sessions
Max 5-hour and weekly limit, plus a separate weekly limit for Fable models Same
Ultra Weekly limit and a separate Fable limit; verified accounts have no 5-hour limit Same
Team, seat-based Enterprise A per-seat allowance on a rolling 5-hour and a weekly window; Premium seats get more Claude chat and Cowork
Usage-based Enterprise No per-seat limit; billed at API rates
Console API key, Bedrock, Agent Platform, Foundry No plan limit; you pay per token, within your organization’s rate limits

Prices per plan are on Claude Code pricing, and what a developer typically spends in tokens is on token cost per developer. This page is about the meter itself.

How to check your usage

  1. /usage is the main screen. /cost and /stats open the same thing. On a Pro, Max, Ultra, Team or Enterprise plan it shows your plan usage bars and a breakdown of what counted against them: shares for skills, subagents, plugins and individual MCP servers, plus recommendations such as trimming a long CLAUDE.md. Press d or w to switch between the last 24 hours and the last 7 days.
  2. The fix shortcut. When the footer says something like f to fix 2 of 3, pressing f lets Claude Code apply the recommendations it can, showing you each change first.
  3. The Session block in /usage shows tokens and an estimated dollar figure at list price. That figure is meant for API users. On a subscription it does not appear on any bill.
  4. /status shows version, model, account and connectivity, which is handy to confirm which account you are on.
  5. /context shows what currently fills the context window. Usage grows with context size, so this is where savings start.

For admins: on Team and Enterprise, the spend report in org analytics lists estimated spend per user and per model once usage credits are on, with CSV export. On the Console there is a cost page, workspace spend limits and a per-member dashboard.

What happens when you hit the limit

The message tells you which ceiling you reached:

  • “You’ve hit your session limit” or “You’ve hit your weekly limit”: the window covers all models, so /model will not help. The message says when it resets.
  • “You’ve hit your Opus limit” or a similar model-specific message: switching to another model family does keep you going.
  • A spend-limit message: you are on usage credits and reached a cap someone set.

Your options then, from the docs:

  1. Wait. On recent versions Claude Code can wait for the reset and continue the interrupted task on its own. /rate-limit-options shows the choices.
  2. Use usage credits. /usage-credits opens the billing settings on Pro, Max and Ultra. Team and Enterprise members without billing access send a request to their admin instead.
  3. Move to API billing. Pro and Max users can switch to a Console account with API credits at standard rates.
  4. Upgrade the plan with /upgrade.

If you run out mid-answer on the 5-hour limit, Claude Code may finish the current step and show “Usage limit reached · wrapping up”. That allowance is rolling out gradually.

The misunderstanding: “I barely typed anything today”

Usage follows context, not keystrokes. Claude Code sends the whole conversation with every request, and every tool call sends another request with its results. A one-line question in a session that has been open all day still pays for the whole history, mostly at the cached rate. The docs list more quiet consumers: the first message after a break that outlasted the cache, /loop and other scheduled tasks firing while you are away, subagents and agent teams with their own context windows, and compaction itself, which reads the whole conversation it summarizes. /clear between unrelated tasks is the cheapest habit there is. Also check whether ANTHROPIC_API_KEY is set: if it is, Claude Code bills that key instead of your plan.

What this means for a team

Seat limits are per person, so a team does not pool its allowance: one developer can be blocked while a colleague has room to spare. The docs advise starting with a small pilot group to set a baseline, and budgeting more for a coding seat than for a chat seat, since one debugging session can consume more than a day of chat. Part of that goes to Claude reading files to learn how a project works, which is why the docs suggest skills that hand Claude that knowledge directly and a CLAUDE.md kept under 200 lines.

Where Bombos fits

Part of a session’s context can go to the agent finding out what a colleague’s agent already found out last week. Bombos is a plugin that gives Claude Code and the team’s other AI tools one shared memory, so team knowledge and learned rules are there for every agent, with their source, instead of each person’s agent working them out again. One developer can try it on their own agents first.

Related questions

Is /cost still a command?

Yes, as an alias. /cost and /stats both open /usage, which shows session cost, plan usage limits and activity stats.

Does switching to a cheaper model help when I hit my limit?

Not for the session or weekly limit, which covers all models. It does help after a model-specific message such as "You've hit your Opus limit": switch to a model outside that family with /model.

Does /usage include what I used on claude.ai or another laptop?

Partly. The plan usage bars reflect your account-wide limits, which Claude chat shares. The detailed breakdown of what counted is computed from session history on this machine, so usage from other devices or claude.ai is not in it.

Can I keep working past the limit?

Yes, if usage credits are on. Run /usage-credits to manage them, or on Team and Enterprise without billing access, to ask your admin. The extra use is billed at API rates.

Do cloud sessions use my limit too?

Yes. Cloud sessions share rate limits with all your other Claude and Claude Code usage, so running tasks in parallel uses your plan faster.

Sources

  1. Claude Code docs: Manage costs effectively
  2. Claude Help Center: Using Claude Code with your Pro or Max plan
  3. Claude Help Center: Use Claude Code with your Team or Enterprise plan
  4. Claude Code docs: Commands
The first half hour · free

What does your AI forget?

  1. 01You tell us which AI you use and where it keeps starting over.
  2. 02We show what a shared memory changes there, on your own examples.
  3. 03You keep the plan, with or without Bombos.
Piet Baudoin
Piet BaudoinYou talk to Piet, from the Bombos teamDutch team from Utrecht and Nijmegen · data in Europe
We reply the same working day.