AI plan calculator

Claude Pro vs Max calculator

Pro is $20 a month, Max 5x is $100 and Max 20x is $200. Max plans now also include $100 or $200 a month of API credits for the Claude Platform. Enter how you use Claude Code to see what that use would cost on the API and what each plan really costs you.

Your inputs

A session is one task from prompt to finished change.
$
What you already spend on the Claude API for your own apps. Max plan credits can pay for it.

Result

Your Claude Code use is worth $150 a month at API prices

That is 7.5 times the Pro price. Max 5x costs $100 and Max 20x $200; their API credits only help if you also build on the API.

Cost per session on Claude Sonnet 5.5
$0.36
Your use at API prices
$150 a month
API only (plus other API spend)
$150 a month
Claude Pro
$20.00 a monthusage limits apply
Claude Max 5x
$100 a month5x Pro usage; $0.00 of the $100 API credits used
Claude Max 20x
$200 a month20x Pro usage; $0.00 of the $200 API credits used

Worked examples

Each row is the calculator run on the inputs shown, with every other input at its default.

UserSessions a dayOther API spendUse at API pricesMax 5x net
A few fixes a day3$0$11.24 a month$100 a month
Daily coding10$0$74.91 a month$100 a month
All-day agent work20$0$150 a month$100 a month
All-day on Opus 5.520$0$300 a month$100 a month
Daily coding plus a $100 API app10$100$74.91 a month$100 a month
Parallel agents plus a $300 API app60$300$449 a month$300 a month

How it is calculated

The calculator prices your Claude Code use the way the API would bill it, then sets that against the flat plan prices.

session cost = size x 1.3 x (cache reads x read price + cache writes x write price + input x input price + output x output price) / 1,000,000 use per month = session cost x sessions per day x active days plan cost = plan price + other API spend - API credits the plan includes (up to that spend)

The typical session is the example Anthropic shows for the /usage command: 940,000 cache read tokens, 50,000 cache write tokens, 1,200 input and 5,300 output tokens, measured on Sonnet 4.6. Claude 4.7 and later models make about 30% more tokens for the same text, so the counts are raised by 1.3.

Max 5x and Max 20x include $100 and $200 a month of API credits for the Claude Platform. If you also build on the API, those credits pay for part of that spend, so the plan's net cost is lower. Credits only count up to what you would spend anyway.

What this page cannot tell you is whether a plan's limits cover your use. Anthropic publishes them only as multiples (Max 5x is five times Pro's usage, Max 20x twenty times), with limits that reset every five hours plus weekly limits. If you keep hitting Pro's limit, that is the signal to move up.

Questions

Is Claude Max worth it?

If your Claude Code use would cost more than $100 a month on the API and you keep hitting Pro limits, Max 5x usually pays off. Max now includes $100 (5x) or $200 (20x) a month of API credits, so if you also build on the API its net cost is lower.

What is the difference between Claude Pro and Max?

Pro is $20 a month ($17 billed yearly). Max 5x is $100 with five times Pro's usage, and Max 20x is $200 with twenty times. Max adds higher output limits, early access, priority at busy times and monthly API credits.

Does Claude Max include API credits?

Yes. Anthropic lists $100 to $200 a month in API credits to build on the Claude Platform with Max, matching the plan price.

How many Claude Code sessions does Pro allow?

Anthropic does not publish a session or token count. Limits reset every five hours and weekly, and depend on model and task size.

Sources

Prices and plans change. This page was last checked on 2026-10-29. The data is open: llm-prices.json, ai-plans.json (CC BY 4.0).

Related calculators