Home › Rules › Claude Pro vs Max vs Ultra

Claude Pro vs Max vs Ultra for Claude Code: prices, usage limits, and which one to pick

From Anthropic's own pricing page, help center and Claude Code docs, checked Nov 1, 2026. Sources below.

• Pricing
• Claude Code
• Usage limits
• Beginners

Short answer: All three paid Claude plans include Claude Code, and they all draw from the same usage pool as your Claude chats. Pro is $20 a month (or $17 a month if you pay $200 for the year). Max is $100 a month for 5x Pro's usage, every 5 hours and every week. Ultra is $200 a month for 10 times Pro's weekly usage, with no 5-hour limit once Anthropic has verified your account. Max and Ultra also include Fable, Anthropic's top model, for up to half your weekly limit, plus monthly API credits ($100 or $200) that you can't spend in Claude Code itself. Max 20x is gone: Anthropic renamed it Ultra and moved its subscribers over with no price change.

The three plans side by side

The Free plan doesn't include Claude Code. Anthropic's help center says the Max and Ultra prices are for web sign-ups and that mobile app prices may be different, and Pro's price is shown in your local currency where it's supported. Opus, Sonnet and Haiku are on all three paid plans. Claude Team plans are separate (a Premium seat gets 5x the usage of a Standard seat) and aren't covered here.

What "5x" and "10 times" actually mean

Anthropic's pricing page says "there's no fixed message count." How far your allowance goes depends on how long your chats are, how big your files are, the model and the effort level. Each plan's limits are measured against Pro:

• The 5-hour limit (Anthropic calls it the session limit). The 5 hours start when you send your first message, and the limit resets in full 5 hours later. If you hit it before then, you wait for the reset. Pro has one, Max has one that's 5x bigger, and verified Ultra accounts don't have one.
• The weekly limit. This covers all models and resets at a fixed day and time each week that Anthropic assigns to your account. It doesn't move based on when you subscribed or when you start working, and you get the full amount each week. Max's weekly limit is 5x Pro's and Ultra's is 10 times Pro's.
• The Fable weekly limit (Max and Ultra only). You can spend up to half of your weekly limit on Fable. It isn't extra: Fable comes out of the same weekly limit and uses it faster than other models. As Anthropic puts it, "you can never use more than your weekly limit."

Everything shares one pool. Claude on the web, the desktop app, the phone apps, and Claude Code in your terminal or your editor (VS Code, Cursor, JetBrains) all count against the same limits. A long chat in the morning leaves less for coding in the afternoon.

Anthropic also says it may limit usage in other ways "to manage capacity" (weekly and monthly caps, or limits on certain models and features). So treat 5x and 10 times as how the plans compare, not as a fixed amount of work.

How much does Pro give you? Anthropic doesn't put a number on it. The pricing page says Pro gives "at least 5x more usage per 5-hour session than Free," and Max and Ultra are measured against Pro.

The math on price

On Anthropic's numbers, the weekly allowance grows at the same rate as the price: $20 for 1x, $100 for 5x, $200 for 10 times. Moving up isn't a bulk discount. What changes is how often you get stopped: Max has a bigger 5-hour limit, Ultra drops the 5-hour limit for verified accounts, and both add Fable and API credits.

Max 5x vs 20x: what happened to Max 20x?

Max used to have a 20x option next to 5x. As of Nov 1, 2026, Anthropic's pricing page has no 20x option. Its help article on the Ultra plan says it plainly: "The Max 20x plan is now called the Ultra plan. Existing Max 20x subscribers move to Ultra automatically, with no price change."

• If you searched "Max 5x vs 20x": today that's Max ($100, 5x Pro every 5 hours and every week) vs Ultra ($200).
• Ultra is described differently: Anthropic now describes it as 10 times Pro's weekly usage with no 5-hour limit for verified accounts, rather than "20x". Anthropic doesn't explain the change in wording, so we won't guess at it. The weekly limit and the missing 5-hour limit are what its pages describe now.
• If you were on Max 20x: you're on Ultra now, at the same price. Check Settings > Usage to see your current limits.

What "verified" means on Ultra

Anthropic verifies Ultra accounts automatically, "using factors like your account and payment history." Until then, a temporary limit applies, and it's removed on its own once you're verified. Verifying your phone number at checkout doesn't remove it. To check: go to Settings > Usage. If there's no "Current session" bar, your account is verified.

Which plan fits which kind of Claude Code user

Based only on how Anthropic describes the limits:

• Pro ($20, or $17 a month paid yearly): fits if you use Claude Code in shorter stretches and can work within a 5-hour limit and a weekly limit. It's the only plan you can pay for yearly. Fable isn't part of Pro's limits, so on Pro you'd pay for Fable with usage credits.
• Max ($100): Anthropic's own advice to Pro users is to consider Max "if you consistently hit limits and need more capacity for larger repositories." You get 5x both limits, Fable for up to half your week, and $100 a month in API credits if you also build your own apps on the Claude API.
• Ultra ($200): Anthropic tells Max users to consider Ultra "if you consistently hit limits." Its pricing card calls it the plan "for your longest tasks." With no 5-hour limit (once verified), you can do a long session in one go and only the weekly limit applies. It also comes with $200 a month in API credits.

A middle option: you don't have to move up a plan to get past a limit now and then. Every paid plan can turn on usage credits and pay API rates only for the extra. If you only hit the limit once in a while, compare what that costs against the price difference before you upgrade.

Switching plans: upgrades take effect right away, with unused time on your old plan credited toward the new one. Downgrades take effect at the end of your billing period.

What happens when you hit a limit

Claude Code warns you as you get close to a limit. Then you have three choices:

1. Wait for the 5-hour or weekly reset.
2. Upgrade: Pro to Max, or Max to Ultra.
3. Turn on usage credits. Your work keeps going at standard API rates instead of stopping, and the message changes to "Now using credits." This works for the 5-hour limit, the weekly limit and the Fable limit.

The wrap-up allowance: if you hit the 5-hour limit while Claude Code is partway through a response, it may keep going briefly to finish that step, and you'll see "Usage limit reached · wrapping up". It's small and capped, it counts toward your weekly limit, and it only finishes the response already running. On Pro it can happen once per weekly period, and on Max each time you reach a 5-hour limit. Anthropic says it's rolling out gradually, so you may not have it yet.

How to turn on usage credits (and cap them)

1. Go to Settings > Usage on claude.ai and find the Usage credits section. (In Claude Code, the /usage-credits command opens the same page in your browser.)
2. Click Enable and add a payment method if you don't have one on file.
3. Click Adjust limit to set a monthly spend limit. You can also pick "Set to unlimited", but a limit is the safer choice.
4. Prepay with Add funds. Auto-reload is optional; it buys more when your balance drops below an amount you choose.

Your usage stops at your monthly spend limit, or when your balance runs out with auto-reload off. You can turn usage credits off any time in the same place. If you subscribed through the iPhone or Android app, you can only turn them on in Claude on the web.

Anthropic also sells prepaid usage bundles at a discount ($50 for $45, $250 for $200, $1,000 for $700), used only after you've gone past your plan's limits.

The monthly API credits don't cover Claude Code

The $100 (Max) or $200 (Ultra) a month in API credits are for building your own apps and agents on the Claude Platform: the Claude API, the Console Playground, Managed Agents and the Agent SDK. Anthropic's help center says they don't apply to interactive Claude Code (terminal, IDE, desktop or web) and can't pay for usage credits. They also don't raise your plan's limits. A few more details:

• You claim them on claude.ai under Settings > Billing, in the API credits section, by linking one Claude Console organization. You can claim after 7 days on the plan.
• They refresh each billing cycle and don't roll over.
• Pro doesn't get them.

Watch for an API key on your computer

If the ANTHROPIC_API_KEY environment variable is set, Claude Code can use that key instead of your plan once you approve it, and you're billed at API rates. To stay on your plan, run unset ANTHROPIC_API_KEY, then run /status in Claude Code to see which sign-in is active. If you're signed in through a Console account, run /login and pick your Pro, Max or Ultra account.

How to check your usage

• On claude.ai or the desktop app: Settings > Usage shows "Current session" (how much of your 5-hour limit you've used and how long until it resets), your weekly limits (all models, plus Fable on Max and Ultra) and when they reset, and your usage-credit spend.
• In Claude Code: run /usage. On a Pro, Max or Ultra plan it shows your plan usage bars, plus a breakdown of what used your limits (skills, subagents, plugins, MCP servers) with suggested fixes. Press d or w to switch between the last day and the last week. Those breakdown figures only count sessions on that computer.
• In VS Code: the Claude Code extension shows the same breakdown in its Account & usage dialog.

How to make your limits last

These come from Anthropic's Claude Code docs and help center, in plain words:

1. Clear between tasks. Every message resends the whole conversation so far. Run /clear when you switch to something unrelated, and /compact when you're in the middle of a long task and want to keep going. Anthropic calls /clear "the single most effective lever for both quality and cost."
2. Match the model to the job. Sonnet for most coding, Opus for hard bugs and big changes, Haiku for quick mechanical edits. Anthropic's help center says Opus "uses meaningfully more of your quota." /model opusplan plans with Opus and builds with Sonnet.
3. Plan before big changes. Press Shift+Tab to switch to plan mode, so Claude proposes an approach before it edits anything. A wrong 400-line change you have to undo costs far more than a plan. Our plan mode guide has plan prompts you can copy.
4. Keep CLAUDE.md short. It's loaded into every session. Anthropic suggests keeping it under about 200 lines and adding a rule only the second time you have to correct Claude on the same thing. See our CLAUDE.md examples, or the short one below.
5. Move one-off instructions into skills. A skill only loads when it's needed, so it doesn't take up space in every session. Starters: Claude Code skills examples.
6. Send noisy work to subagents. Running tests or reading long logs fills your context. A subagent keeps that output in its own context and sends back a summary. It still uses your limits, so give simple subagents a smaller model. See Claude Code subagents examples.
7. Filter output with a hook. Anthropic's docs show a hook that cuts test output down to just the failures before Claude reads it. More in our Claude Code hooks examples.
8. Point at files instead of pasting them. Write the path, like src/auth.ts, and Claude reads only what it needs.
9. Turn off what you don't use. Run /mcp to switch off MCP servers you're not using, and lower the effort level with /effort for simple tasks.

A short CLAUDE.md

Keep it about your real project. The "Compact instructions" part at the end is a tip from Anthropic's docs: it tells Claude what to keep when it summarizes a long session.

# CLAUDE.md

## Project
- A small web app for [who it's for] that [the main thing it does].
- Stack: [for example Next.js + Supabase].

## Commands
- Run locally: npm run dev
- Tests: npm test
- Lint: npm run lint

## Rules
- Read only the files you need for the task.
- Make one small change at a time, then tell me how to test it.

# Compact instructions
When you compact, keep the test output and the code changes.

Build one for your exact project in the free rules generator, or see AGENTS.md vs CLAUDE.md if you also use other tools.

How this compares to Cursor, Codex and Copilot

Cursor, OpenAI Codex and GitHub Copilot each count usage their own way. For their prices and limits next to Claude Code's, see Cursor vs Claude Code vs Codex and GitHub Copilot vs Cursor vs Claude Code.

Related

• Cursor vs Claude Code vs Codex (real prices)
• GitHub Copilot vs Cursor vs Claude Code (prices side by side)
• CLAUDE.md examples
• Plan mode guide (6 tools + 6 plan prompts)
• Claude Code subagents examples
• Claude Code skills examples
• Claude Code hooks examples
• Lovable vs Claude Code
• GitHub Copilot AI credits explained

Sources

• Anthropic: Claude pricing (plans and FAQ)
• Claude Help Center: Use Claude Code with your Pro, Max, or Ultra plan
• Claude Help Center: What is the Pro plan?
• Claude Help Center: What is the Max plan?
• Claude Help Center: What is the Ultra plan?
• Claude Help Center: How do usage and length limits work?
• Claude Help Center: Manage usage credits for paid Claude plans
• Claude Help Center: Claude Fable models on your plan
• Claude Help Center: Monthly API credits for Max, Ultra, and Team plans
• Claude Help Center: Buy usage bundles
• Claude Help Center: Models, usage, and limits in Claude Code
• Claude Code docs: Manage costs effectively
• Claude Code docs: Authentication

All prices, limits and settings checked Nov 1, 2026 on Anthropic's own pricing page, help center and Claude Code docs. Prices are in US dollars for web sign-ups, before tax. Anthropic says prices, plans and usage limits can change. If something here no longer matches, tell us at [email protected].