Claude limits hub

Claude limits, resets, and upgrade decisions in one place

If Claude keeps cutting you off, the real question is not just how many messages you get. It is which limit you are hitting, what is burning the window, and whether a plan upgrade or a workflow fix will solve it.

Recommended trail

The smartest path from limit warning to fix

Diagnose the limit first, compare the plan only when needed, then keep a live meter in your Mac menu bar so the same problem does not surprise you again.

See all apps

Diagnose the limit

Work out whether you are hitting the 5-hour window, weekly cap, Claude Code cap, or API budget.

Open page

Compare the upgrade

Check Claude against ChatGPT, Google AI, Perplexity, X Subscription and Microsoft 365 before buying more capacity.

Open page

Tokens 4 Breakfast

Private macOS menu bar tracker for AI spend, token usage, Claude limits, Cursor, Copilot, OpenAI, Gemini, and more.

Open Tokens 4 Breakfast

Start here

Claude limits explained

The 5-hour window, weekly cap, reset behavior, and the workflow habits that stretch each session.

Open

Claude Code

Claude Code rate limit hit?

What is actually happening when Claude Code slows down or stops, and how to forecast the weekly cap.

Open

Upgrade decision

Claude Max vs ChatGPT Pro

Max 5x and 20x against ChatGPT Pro $100, $200 and $500, Google AI Ultra and X Super, with Germany pricing notes.

Open

Cross-vendor

AI usage limits compared

How Claude’s windows compare with ChatGPT, Gemini, Copilot, Cursor and Perplexity, from vendor documentation only.

Open

Claude limits by plan

Anthropic publishes windows and multipliers, not message counts: its pricing page says there is no fixed message count. Claude Code uses the same 5-hour and weekly pool as chat, and the Max API credits cannot be spent in Claude Code or the Claude apps. Last verified October 23, 2026; prices are USD and exclude tax, even for buyers in Germany.

Plan	Price (USD)	5-hour session	Weekly cap	Notes
Free	$0	Rolling 5-hour window	Not stated	No Opus, no Fable, no Claude Code
Pro	$20/mo, or $17/mo billed annually	At least 5x Free per 5-hour session	Weekly limit across all models, reset at a fixed time per account	Fable 5.1 through usage credits; Claude Code wrap-up allowance once per week
Max 5x	$100/mo, monthly only	5x Pro per 5-hour session	Weekly limit across all models plus a separate weekly Fable limit	Fable up to 50% of weekly limits; $100/mo API credits for the Claude Platform
Max 20x	$200/mo, monthly only	20x Pro per 5-hour session	Weekly limit plus a separate weekly Fable limit	$200/mo API credits for the Claude Platform
Team Standard	$20/seat/mo annual or $25 monthly	1.25x Pro per session, per member	Weekly limits, not pooled	Up to $500/mo pooled API credits per team
Team Premium	$100/seat/mo annual or $125 monthly	6.25x Pro per session, per member	Weekly limits, not pooled	For the heaviest users on a team

Claude pricing Max plan help Team plan help Claude Code with Pro or Max

What to do next

If you are hitting limits during chat work

Start with shorter task-specific chats, Projects for reusable context, and lighter tools for low-stakes work. Then decide whether Pro or Max is worth it.

If Claude Code is the bottleneck

Watch the 5-hour and weekly usage pattern. Coding agents burn limits differently, so compare Claude Code against Cursor, Copilot, and Codex by real workflow cost.