Claude Usage Limits Explained: The Essential Guide to 5-Hour and Weekly Caps

Claude usage limits are the thing everyone complains about and almost nobody can explain, mostly because Anthropic never gives a message count. So here’s what the official pages actually say, checked on October 21, 2026: the pricing page FAQ, the help center, and the Claude Code docs. Where there’s no official number, I’ll tell you there isn’t one.

Quick heads-up before anything else: the free limit reset Anthropic handed out on September 22 must be used by October 22, 2026. If you’re on Pro, Max or Team, look in Settings → Usage. More on that in the changes log below.

Claude usage limits in one paragraph

Every plan has a limit that resets on a rolling five-hour window. Paid plans add a weekly limit on top. Everything you do in Claude, on the web, the desktop app, your phone, and Claude Code, draws from the same pool. How fast you burn through it depends on how long your conversations are, which model you pick, and which features you use. That’s why there’s no fixed number of messages. Anthropic says this outright, and honestly I wish they’d at least give a range.

What each plan gets

Free is the baseline. It covers everyday questions and doesn’t include Claude Code.

Pro is $20 a month, or $17 a month on the annual plan ($200 paid up front). The pricing FAQ says Pro gives you at least 5x the usage of Free per five-hour session. It includes Claude Code.

Max starts at $100 a month for 5x the usage of Pro per session, or $200 for 20x. Monthly billing only. Max also comes with $100 or $200 a month in API credits (more on why that’s not “more Claude Code” below), and the plan table lists Fable at “50% of weekly limits” for both Max tiers, while on Pro Fable runs on usage credits.

Team has Standard seats ($20 per seat per month annual, $25 monthly) with more usage than Pro, and Premium seats ($100 annual, $125 monthly) with 5x Standard. Anthropic also reserves the right to add other caps, weekly or monthly, model or feature-specific, “at our discretion”. That line is in the FAQ. It’s vague, and I don’t love it.

Should you pay for Max? That’s its own argument; mine is in Claude Pro vs Max.

The two clocks: five-hour session and weekly limit

This is the part that trips people up. Your usage counts against the session window and the weekly limit at the same time. The session window rolls every five hours. The weekly one resets at a fixed day and time assigned to your account, which you can see in Settings → Usage. The Claude Code error docs put it plainly: one heavy burst, like a big parallel workflow, can use up the weekly allowance before the session window even resets.

Switching models doesn’t refill either clock; both are shared across models. The exception is the model-specific Opus or Sonnet limit. If you hit one of those, /model to a model in another family and keep going.

How to check how much you have left

On the web or desktop: Settings → Usage. In Claude Code: /usage (on paid plans it shows the plan bars, when each resets, and a breakdown of what’s eating into them, like skills, subagents and MCP servers; see costs docs). Claude Code also warns you before you run out, with something like You've used 85% of your session limit · resets 3:45pm. If you like dashboards, you can put the rate limit fields in a custom status line.

I don’t trust third-party “quota radar” sites for this. They can’t see your account. The official screens can.

What happens when you hit the limit

Three options, per the FAQ: wait for the reset, move to a bigger plan, or turn on usage credits on a paid plan and keep working at standard API rates (extra usage help article). In Claude Code that’s /usage-credits. Since v2.1.234, Claude Code can also just wait in the open session and continue the task after the reset; the toggle is “Continue automatically at usage limit” in /config. The exact messages and what each one means are in the Claude Code errors lookup.

A small thing that saves arguments: a 529 Overloaded error is Anthropic’s capacity, not your quota, and it doesn’t count against your limits.

Max API credits are not extra Claude Code

This confused me at first too. The $100 or $200 a month on Max is credit for the Claude Platform API. According to the help article and the platform docs, it doesn’t apply to interactive Claude Code under your subscription or to extra usage, it doesn’t roll over, and you claim it by linking a Console organization once you’ve been on the plan for 7 days. Team gets up to $500 a month, pooled. Useful if you build stuff on the API. Not a bigger Claude Code allowance.

Claude usage limits changes log (2026)

Only changes with an official Anthropic source. I’ll keep adding to this. The broader limit-and-ban history for Claude and Codex lives in my usage limit tracker.

March 14 (US time). The official @claudeai account posted: “We’re doubling usage outside our peak hours for the next two weeks” (post). The help page for that promo now returns 404, so the post is the only official source I can still point to. It’s over.

May 6. Anthropic announced it was doubling Claude Code’s five-hour rate limits for Pro, Max, Team and seat-based Enterprise, and removing the peak-hours reduction on Claude Code for Pro and Max. No end date was given.

September 22. With Opus 5.5, five-hour limits in Claude Code went up 20%, and since Opus 5.5 is priced lower it goes about 25% further, per @ClaudeDevs. Pro, Max and Team also got a saved reset, available in Settings → Usage “any time until Oct 22” (post). The limit reset help page says the “Reset for free” button lives in Settings → Usage, isn’t available in the mobile app or in Claude Code’s terminal or IDE, and shows when it expires. The posts don’t name a time zone for the deadline, so don’t leave it to the last hour.

Claude usage limits FAQ

How many messages do I get on Pro? Anthropic doesn’t publish a number. It’s at least 5x Free per five-hour session, and the real amount depends on conversation length, model and features.

Did Claude double usage limits? Three times this year, in different ways: off-peak doubling for two weeks in March (ended), Claude Code five-hour limits doubled on May 6, and +20% plus a free reset on September 22. See the log above.

Is there a usage limits increase I can buy? Max 5x or 20x, or usage credits at API rates on any paid plan.

Does Claude Code have separate limits? No. Claude Code shares the same pool as chat, so a long coding session eats into your chat allowance and vice versa.

What about Reddit threads saying limits got cut? I can’t use those as a source. If there’s an official change, it goes in the log above with a link.

How does this compare with ChatGPT and Gemini? I lined up the quota rules side by side in AI subscription quota rules.

More on this: Claude and Codex usage limit tracker · AI subscription and API price comparison · Claude Code errors lookup · Claude category