Zest

CLAUDE · PRICING

Is Claude Max worth it?
Our month's math.

Yes, if Pro's limits stop you most weeks or you'd buy $80+ of credits a month.

Start free
npx @winding-labs/zest

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

Flyps Charm Wildfires Geoforge MisaLabs

Is Claude Max worth it?

Claude Max is worth it for anyone who runs Claude Code most days; for occasional chat and short sessions, Pro is enough. Per unit of usage, Max costs the same as Pro: $100 for five times Pro's $20 allowance. What you pay for is the headroom, Fable 5.2 inside the plan, priority at busy times and $100 a month in API credits. Max is worth it if most of these are true:

  • Pro's five-hour limit stops you most days, or its weekly limit most weeks.
  • You'd spend more than $80 a month on usage credits to keep going on Pro.
  • You run agents or subagents that keep working while you do something else.
  • You want Fable 5.2 without paying for it per token.

When does Claude Max cost less than Pro plus usage credits?

Once your usage credits on Pro would pass $80 in a month. At our own month's average Claude Code response on Opus 5.5, $80 of credits runs about 1,136 responses, or 11 sessions like our median one ($7.25). Claude Code also caches for five minutes instead of an hour on credits, so a paused session costs more to pick up. Try your own month:

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.
PlanPrice a monthCosts less than the API after
Pro$202 days
Max$1008 days
Ultra$20016 days

How much does Claude Max cost, and what replaced Max 20x?

Claude Max costs $100 a month, billed monthly only, and the $200 Max 20x is now called Ultra. Ultra gives 10x Pro's weekly usage and no five-hour limit for verified accounts; Max 20x subscribers moved over at no extra cost. Claude Max plan has the full plan.

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 2026NowPrice
ProPro$20 a month
Max 5xMax$100 a month
Max 20xUltra$200 a month

How much usage do you get with Claude Max?

Five times Pro's usage every five hours and every week; Anthropic publishes no token or message count for any plan. What fits depends on message and file length, how long the conversation is, the model, the effort level and tools such as web search. Fable 5.2 can use up to half the weekly limit and uses it faster than Opus.

How do Max's limits reset, and do unused windows roll over?

The five-hour window starts with your first message and resets in full five hours later; the weekly limit resets at a fixed day and time assigned to your account, shown in Settings > Usage. Each window resets to full rather than adding to the next, so usage you don't spend doesn't carry over. The limits are shared across Claude, Claude Code and Cowork.

What does a heavy Claude Code month cost at API rates?

Far more than any plan: Our own month: 2,228 Claude Code sessions on the Mac that runs our coding agents, October 22 to November 20, 2026, each response counted once and priced at Anthropic's list prices. Every session ran signed in to a Claude plan, so this is what the same work would cost on an API key, not our bill. It is a heavy, agent-run month: 72% of responses came from subagents.

At $63,936 for the month, that is 639 times Max's price. Usage-limit notices showed on 16 of 30 days. A developer running one agent at a time sits far below this, nearer Anthropic's $150 to $250 a month.

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.
MeasuredValue
Sessions and API responses2,228 sessions, 874,770 responses
Tokens158.8 billion, 97.6% of them cache reads
ModelOpus 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 wentCache 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 notice16 of 30

Does Max give more Opus and Fable than Pro?

Yes: five times the Opus, since Opus shares the plan's limit, and Fable 5.2 inside the plan for up to half the weekly limit, where Pro pays for every Fable token with usage credits. Fable costs twice Opus per token on the API, so on Pro it is where usage credits go fastest.

Do Max's API credits pay for Claude Code?

No: Max's $100 a month in API credits doesn't cover interactive Claude Code or usage credits. It covers claude -p and the Agent SDK only when you run them with an API key from the linked Console organization; signed in with your plan, they draw on the plan's limits. So Max against "$100 of API credits" is no contest for Claude Code: a subscription inside its limits covers far more than $100 of usage at list price.

What the monthly API credits on Max, Ultra and Team pay for, from Anthropic's Help Center. Checked November 21, 2026.
Pays forDoesn't pay for
The Claude API (Messages and Message Batches)Interactive Claude Code in the terminal, IDE, desktop or web
The Playground in the Claude ConsoleUsage credits (extra usage) in Claude, Claude Code or Cowork
Claude Managed AgentsClaude on Amazon Bedrock, Google Cloud Vertex AI or Microsoft Foundry
The Claude Agent SDK, and claude -p run with an API keyThird-party apps billed through linked billing

When is Pro enough, and who should skip Max?

Skip Max if you use Claude for chat and writing, run Claude Code a few times a day, or hit Pro's limit less than once a week. If you don't code, Max pays off only for long research, document and Cowork tasks that run out Pro's window: tasks Claude keeps running in the cloud, scheduled ones included. Pairing Pro with another $20 chat plan spreads work across two allowances, at the cost of two tools. Claude Pro vs Max sets the two plans side by side.

Is Claude Ultra worth it over Max?

Ultra is worth it when the five-hour window, not the week, is what stops you: for $100 more it drops that limit once verified and doubles Max's weekly usage. If you hit Max's weekly limit rarely and the five-hour one only on long days, stay on Max. Claude Ultra plan compares them.

Claude Max and Ultra side by side, from claude.com/pricing and Anthropic's Help Center. Checked November 21, 2026.
MaxUltra
Price$100 a month$200 a month
Five-hour limit5x ProNone for verified accounts
Weekly limit5x Pro10x Pro
Fable 5.2Up to half the weekly limitUp to half the weekly limit
Monthly API credits$100$200
Price per Pro-sized week of usage$20$20

What happens when you hit Max's limit mid-project?

Claude Code finishes the step it is on, then waits for the reset, or keeps going on usage credits if you turned them on; it can resume by itself when the window reopens. These are the notices it showed in our own logs:

Limit notices in our Claude Code logs, October 22 to November 20, 2026
You've hit your 5-hour limit · resets 3:50pm (America/Los_Angeles)
You've hit your weekly limit · resets Nov 20 at 7am (America/Los_Angeles)
Usage limit reached · continuing automatically at 5pm · esc to cancel
Usage limit reset · continuing automatically

How do you make Max last longer?

Anthropic's usage tips, and two from our own month:

  • Keep reference files in a Project: cached content counts less against the limit.
  • Batch related questions into one message, and start a new conversation for a new task.
  • Save Fable and Opus for the work that needs them; Sonnet 5.5 handles routine edits.
  • Lower the effort level for simple tasks.
  • Watch subagents: they were 72% of our responses, and each warms its own cache for five minutes.
  • Schedule long tasks for windows you would otherwise leave idle.

What do Reddit and Hacker News users say about Max?

Users who code all day call it worth the money; doubters ask whether the price can last. In one Hacker News thread a Max subscriber who runs a business wrote that they'd pay up to $500 a month for the time it saves, and noted the math differs for someone paying from their own salary; replies questioned whether subscriptions cover the cost of serving them. Neither Anthropic nor anyone outside it publishes that cost, so treat today's prices as today's.

What should you check in your first week on Max?

Five checks, so the upgrade proves itself:

  • Settings > Usage: note when your weekly limit resets.
  • Claude Code: run /usage to see what uses your limit, by skill, subagent, plugin and MCP server.
  • Unset ANTHROPIC_API_KEY, or Claude Code bills that key instead of your plan.
  • Claim the monthly API credits after seven days, if you build on the API.
  • Count the days you hit a limit; if it is still most days, look at Ultra.

How do you know Max paid off?

Compare what your sessions would cost at API rates with what you paid. Claude Code's /usage prices one session; on a plan it is a list-price estimate, not a bill.

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

Start free
npx @winding-labs/zest

A FEW THINGS TO KNOW

Questions

Can you get Claude Max free or discounted?+

No. Anthropic offers no standing discount or trial for Max; it runs occasional limited-time promotions.

What mistakes do people make before buying Max?+

Three common ones: skipping Pro, so they never learn whether its limits actually stop them; not tracking usage, so they can't tell whether the upgrade paid off; and in Claude Code, leaving ANTHROPIC_API_KEY set, which bills that key instead of the plan.

Is Claude Max worth it if I don't code?+

Only if long research, writing or Cowork tasks run out Pro's five-hour window most days. Otherwise Pro is enough.

How much more do you get on Pro than on Free?+

At least five times Free's usage per five-hour session, plus Opus and Claude Code.

Is Claude Max priced below cost?+

Nobody outside Anthropic knows. At list prices heavy agent use costs many times the plan: our own month came to $63,936. Anthropic says prices and plans may change.