Free for 3 seats · No credit card · One command signs in every agent on your machine
Sample data
Spend in the Zest app: your agent usage at list prices, by week and by model
USED BY ENGINEERING TEAMS AT
What is the difference between Claude Pro and Max?
Max gives five times Pro's usage in both the five-hour and the weekly window, for five times the price. On top, Max includes Fable 5.2 for up to half the weekly limit (Pro pays for Fable with usage credits), priority at high-traffic times, early access to advanced features and $100 a month in API credits. Pro bills monthly or yearly, Max monthly only.
Claude's individual plans, from claude.com/pricing and Anthropic's Help Center. Checked November 21, 2026.
Pro
Max
Ultra
Price
$20 a month, or $17 billed yearly
$100 a month
$200 a month
Five-hour limit
1x (the base)
5x Pro
None for verified accounts
Weekly limit
1x (the base)
5x Pro
10x Pro
Fable 5.2
Usage credits only
Up to half the weekly limit
Up to half the weekly limit
Opus, Sonnet, Haiku
Yes
Yes
Yes
Claude Code
Yes
Yes
Yes
Monthly API credits
None
$100
$200
Usage credits past the limits
Yes, at API rates
Yes, at API rates
Yes, at API rates
Priority at high-traffic times
Pricing page: no; Pro's Help article: yes
Yes
Yes
Billing
Monthly or yearly
Monthly only
Monthly only
Which costs less for your month, Pro or Max?
Pro costs less until its limits stop you; then every extra hour bills as usage credits at API rates, and Max costs less once those credits pass $80 a month. Enter your own month below.
WORK IT OUT
Enter how many days a month you code with Claude and what one of those days would cost on an API key.
At API rates that month costs $260. Every plan costs less than paying per token: pick the one whose limits hold your week. Max is 5x Pro; Ultra is 10x Pro a week, with no five-hour limit once verified.
Each plan's price against the same days on an API key. A plan saves that only while its limits hold your work.
Plan
Price a month
Costs less than the API after
Pro
$20
2 days
Max
$100
8 days
Ultra
$200
16 days
How much more Claude Code does Max give than Pro?
Five times as much in every five-hour window and every week; Anthropic gives no hours or token counts for either. Max also lets Claude Code finish the step it is on at every five-hour limit, where Pro allows that once a week. How many hours that is depends on the model, the size of the codebase in context and how many subagents run. For scale, in our own month the median Claude Code session came to $7.25 at list price, and one in ten passed $69:
Our Claude Code month, October 22 to November 20, 2026: 2,228 sessions on the Mac that runs our coding agents, priced at Anthropic's API list prices. List price, not our bill. Checked November 21, 2026.
Measured
Value
Sessions and API responses
2,228 sessions, 874,770 responses
Tokens
158.8 billion, 97.6% of them cache reads
Model
Opus 5.5 for 99.1% of tokens
Cost at API list prices
$63,936
The same tokens with no caching
$647,366
Where the cost went
Cache reads 48%, output 24%, cache writes 28%
Median session
$7.25 (59% of sessions under $10)
Median day
$1,806
Days with a usage-limit notice
16 of 30
What does a year of Pro or Max cost?
A year of Pro costs $200 billed up front or $240 monthly; a year of Max costs $1,200, since Max has no yearly price.
What a year of each individual Claude plan costs, from claude.com/pricing, before tax. Checked November 21, 2026.
Plan
Per month
Per year
Pro, billed monthly
$20
$240
Pro, billed yearly
$17 ($200 up front)
$200
Max (monthly only)
$100
$1,200
Ultra (monthly only)
$200
$2,400
When is Claude Pro enough?
Pro is enough when:
You chat, write and run Claude Code a few times a day, not all day.
You hit the five-hour limit less than once a week.
You don't need Fable 5.2 often, or pay for it in usage credits when you do.
Your usage credits stay under $80 a month.
Who should buy Claude Max?
Max fits when:
Claude Code runs most of your working day, or fans work out to subagents.
Pro's five-hour limit stops you most days, or its weekly limit most weeks.
You want Fable 5.2 inside the plan.
You'd use $100 a month of API credits to build on the Claude Platform anyway.
Is Max worth it over Pro?
For daily Claude Code work, usually yes. Anthropic's average Claude Code developer uses about $13 of API-rate usage an active day, 8 days' worth to pass Max's price; Is Claude Max worth it? works through the break-even with our own month.
What about paying per token instead of Pro or Max?
An API key in Claude Code bills every token at list price, with no five-hour or weekly limit, only the Console's spend caps. It costs less than Pro only for very light use, about 2 average days a month. Claude pricing has the per-token prices.
Could something cheaper than Max fit?
Three options cost less than Max for some people. Pro plus usage credits, for occasional overruns. An API key, for light or scripted use: Sonnet 5.5 costs half of Opus 5.5 per token and Haiku 5.5 a fortieth. And for two or more people, Team Standard seats at $25 a month ($20 billed yearly) with more usage than Pro each.
What should you check before upgrading to Max?
Five things that catch people:
Max bills monthly only, so a yearly Pro plan's remaining balance becomes account credit, and only with the same billing address.
Max's API credits don't pay for interactive Claude Code or usage credits.
Through Google Play, an upgrade charges the full price and turns unused time into extra days.
If ANTHROPIC_API_KEY is set in your shell, Claude Code bills that key instead of your plan.
Limits are shared: chat on the web uses the same allowance as Claude Code.
What happened to Max 5x and Max 20x?
Max 5x is now Max at the same $100, and Max 20x is now Ultra at $200: 10x Pro's weekly usage and no five-hour limit for verified accounts. Our price check saw the change on November 3, 2026. Claude Ultra plan compares the top two.
Claude's plan names before and after Ultra, from our checks of Anthropic's pricing pages on October 4 and November 3, 2026, and its Ultra article. Checked November 21, 2026.
Before November 2026
Now
Price
Pro
Pro
$20 a month
Max 5x
Max
$100 a month
Max 20x
Ultra
$200 a month
Claude Max vs ChatGPT Pro: what do you get for $100?
Both start at $100 a month and both include a coding agent: Claude Code on Max, Codex on ChatGPT Pro, which OpenAI sells at $100, $200 or $500 with no five-hour limit for now (read November 18, 2026). Claude Code vs Codex compares the two agents.
How do you make Pro's limits last longer?
Anthropic's usage tips, which our own month bears out:
Start a new conversation for a new task: a long one resends its history on every turn.
Put reference files in a Project: cached content counts less when reused.
Batch related questions into one message.
Use Sonnet 5.5 or Haiku 5.5, or a lower effort level, for routine work.
Turn off connectors and web search a task doesn't need.
How do you see which teammates need Max?
A lead can't see a teammate's Settings > Usage, and a plan's usage carries no price.
Zest prices every Claude Code session it captures at list rates, inside a plan's allowance too, per engineer, and ties each session to the pull request it led to. Where Claude Code sends the limit meters, on Pro and Max, Zest keeps them with each session, so a lead sees who keeps running out of the five-hour window and sizes each seat to the person.
TRY IT ON YOUR TEAM
See what your team ships with AI.
Starter is free (3 seats, no credit card). Pro is $49 a month with 5 seats and Max is $149 a month with 10 seats — priced per workspace, not per engineer, with extra seats at $19 a month. See pricing
Pro gives at least 5x Free's usage per five-hour session, adds Opus and Claude Code, and keeps working past its limits on usage credits.
Can I go back from Max to Pro?+
Yes. A downgrade takes effect at the end of the current billing period, and your chats, projects and files stay.
Is there a discount on Claude Max?+
No. Anthropic has no standing discount on individual plans; it runs occasional limited-time promotions.
Which Claude model is best on each plan?+
Fable 5.2 is Anthropic's top model: inside Max's limits, on usage credits only on Pro. Opus 5.5, which Anthropic positions for long-running agentic coding, is on both.