Your AI plan has two clocks: how to stop hitting Claude usage limits mid-task

How do Claude usage limits work?

Claude's help page on usage limits says paid plans have two kinds: "a session limit that resets every five hours, and weekly limits that reset once every week."

The session limit starts when you send your first message. Five hours later it resets in full, whether you used a little or a lot. Hit it early and you wait out the rest of those five hours.

The weekly limit covers all models and resets at a fixed day and time assigned to your account. That time doesn't move, no matter when you signed up or when you started using Claude this week. Some plans also have a separate weekly limit for Fable, Anthropic's top model.

Everything counts toward the same pool. Chatting on claude.ai, working in Claude Code (Anthropic's coding agent, which writes and runs code for you) and using the Claude Desktop app all draw from one allowance. So an afternoon of coding can leave less room for writing in the evening.

What changed on the Ultra plan?

On October 30, 2026, Anthropic renamed its Max 20x plan to Ultra. According to Claude's pricing page, the plans now line up like this:

• Pro (US$20 a month, or US$17 a month billed yearly): the base amount, every 5 hours and every week.
• Max (US$100 a month): 5x Pro usage every 5 hours and every week.
• Ultra (US$200 a month): 10x Pro usage every week, and no 5-hour limit for verified accounts.

Anthropic says it verifies Ultra accounts automatically, using things like account and payment history. To check yours, the Ultra help page says to open Settings > Usage. If you don't see a "Current session" bar, you're verified.

Dropping the five-hour cap sounds like pure good news, and for people who work in long bursts it mostly is. The catch is that the weekly limit is now the only thing slowing you down, and it arrives all at once.

"One reminder: the weekly limit didn't go up. Without the 5-hour gate, it's easy to get absorbed and burn a whole week of Max 20 usage in a day. Now you have to watch your own usage." — XTian (@XiaoTian258) on X

The change also confused some new subscribers. Mike Overell, who says he leads product for Claude subscriptions, posted on X: "I get that some new users assumed 20x applied to both limits. We could have done better here, and I'm sorry." (source). Ultra's 10x applies to the weekly limit only, as the pricing table shows.

Where can I see how much Claude usage I have left?

Go to Settings > Usage. On plans with a session limit, a "Current session" bar shows how much you've used and how long until it resets. The same page shows when your weekly limit resets next.

We'd make it a habit to look before any job that will run a while, such as a long Claude Code task, a big research request or a chat with lots of files attached. Starting a three-hour job with 10% of your week left is how work stops halfway.

How do I make my Claude usage last longer?

Anthropic lists what drives usage: how long and complex your chats are, which features you use, which model you pick and the effort level you set. Its help page suggests several ways to use less:

• Match the model to the job. Quick questions and simple rewrites rarely need the biggest model. Keep Opus or Fable for the work that needs them.
• Lower the effort level for routine tasks. Anthropic says higher effort uses more tokens, the chunks of text AI usage is measured in.
• Turn off extended thinking when you don't need Claude's deeper reasoning.
• Switch off connected apps and tools a chat doesn't need, and ask Claude not to search the web when current information doesn't matter. Anthropic calls tools and connectors "token-intensive."
• Start a fresh chat for a new task. Very long chats eventually trigger automatic summarizing of older messages, and Anthropic says those chats use more of your limit.
• Keep projects lean. Short project instructions and fewer unused files mean less for Claude to read each time.

On a plan with a weekly limit, pacing matters too. If your reset lands on a Monday, a heavy Thursday leaves you thin for the weekend. Big batch jobs fit best early in your cycle, when you can see how much they cost before the week runs low.

What happens when you hit a Claude usage limit?

You get three choices, per Anthropic: wait for the reset, upgrade, or turn on usage credits. An upgrade applies right away. Usage credits, available on paid plans, let your work keep going at standard API rates instead of stopping, so set a spending cap you're comfortable with before you switch them on.

Upgrading only makes sense if you hit limits often. Hitting the weekly limit once in a busy month is usually cheaper to fix with a few credits than with an extra US$80 or more a month.

Do ChatGPT and other AI plans work the same way?

The pattern is spreading. ChatGPT and Codex, OpenAI's coding agent, will start sharing one allowance in December, so a big coding session there will eat into your chat time the same way it does on Claude. Whatever plan you pay for, find its usage page now, before a deadline makes you look for it.

We think usage meters are fine as long as they're easy to read and you check them before a task, not after it fails. Anthropic's Settings > Usage page does that job. The rest is on us as users: pick the right model, keep chats short and look at the meter first.