How much Claude Opus 5 do you get on Pro and Max in the Claude apps?
Anthropic publishes no Opus 5 message count. Pro is $20 a month, Max 5x $100 and Max 20x $200, sessions reset every five hours, and Opus carries a weekly limit of its own.
Checked against the lab's own pages on September 18, 2026. Limits change often, so the date matters.
People keep asking what one Opus 5 prompt costs them, usually after watching a single reply eat most of a session. Anthropic's help center won't answer that in messages, and it says why.
The Claude apps meter usage rather than counting messages. What you get depends on your plan, and Opus carries a weekly limit of its own on top of everything else.
- Pro
- $20 per month
- Max 5x
- $100 per month, five times the Pro plan's per-session usage allowance
- Max 20x
- $200 per month, 20 times the Pro plan's per-session usage allowance
- Session limit
- Resets every five hours
- Weekly limits
- One for Opus only, one for all other models
- Where to check
- Settings, then Usage
What resets, and when
Your session limit resets every five hours, on Pro and on both Max tiers. Both plans also carry a weekly usage limit across all models, and that one resets at a fixed time each week assigned to your account. The day and time stay put whatever you do, and you get the full weekly allowance each cycle.
Under Settings, then Usage, you'll find progress bars for the current session and for the weekly limits (the same panel shows how much time is left in the session). The weekly section splits in two: Opus only, and all other models. That split is the closest thing to an Opus-specific number Anthropic publishes anywhere.
What counts against it
All of it. Usage on claude.ai, Claude Code and Claude Desktop draws on the same limit, so a heavy afternoon in the terminal shows up in the app.
Message length moves it, and so do attachment size, how long the current conversation has run, which model you picked, the effort level, artifacts, and multi-step work like running code or browsing (that list is Anthropic's own, and it runs to eight items). Turns out caching cuts the other way: content in projects is cached and doesn't count against your limits when reused, and prompts you send often are partially cached.
Long chats are the expensive ones. With code execution enabled, Claude summarizes earlier messages when a conversation nears the context window, which keeps it going and costs you more.
Longer conversations that trigger automatic context management consume more of your usage limit.
What happens when you run out
You can wait for the reset or move up a plan. Pro, Max, Team and seat-based Enterprise subscribers can also turn on usage credits, which keep you working past the usage the plan includes.
Why nobody publishes a message count
Because it would be wrong for most people. Anthropic's own wording is that the number of messages you can send varies with message length, the files you attach, the length of your current conversation, and the model or feature you use. Two people on the same plan doing different work get different numbers, and so do you on two different days.
Anthropic also says it may limit usage in other ways, such as weekly and monthly caps or model and feature usage, at its discretion, and that prices and plans can change the same way. Granted, that line sits on most help pages.