Claude pricing splits into two models as of this writing: per-seat subscription plans — Free, Pro, and Max for individuals, Team and Enterprise for organizations — and usage-based API pricing billed by tokens for teams building Claude into their own software. Everything on an Anthropic invoice traces back to one of those two structures.
You won't find dollar figures in this post, and that's deliberate. Anthropic adjusts pricing, introduces tiers, and evolves the lineup — any number we printed would eventually be wrong, and a stale price is worse than no price. What doesn't go stale is the model: what each tier is for, how the billing works, and what drives cost up. As an Anthropic partner that licenses and deploys Claude for businesses, that's the part we scope every week.
How does Claude pricing actually work?
Two different products, two different meters. The subscription plans work like any modern SaaS: a set rate per person, with higher tiers unlocking more usage capacity and more features. You're paying for a person's access to Claude through the apps — web, desktop, mobile, and tools like Claude Code.
The API is a different animal: you pay for what you consume, measured in tokens — roughly, chunks of the text going in and coming out. No seats, no per-user math; a script that runs once a day and a product serving a million customers are billed on the same meter, just at wildly different volumes. Model choice matters here too, since more capable models carry higher per-token rates than faster, lighter ones.
What are the individual plans — Free, Pro, and Max?
Three tiers, one clear ladder:
Free
Real access to Claude at no cost, with usage limits that make it a genuine trial rather than a working tool. Perfect for figuring out whether Claude fits how you think — most people know within a week.
Pro
The everyday tier for professionals: substantially higher usage limits and access to more capable models and features. If Claude is part of your daily work — writing, analysis, coding, research — this is where you likely land.
Max
For heavy users who hit Pro's ceilings: think developers running long Claude Code sessions or analysts working with Claude most of the day. Max exists because a meaningful group of people use Claude as their primary work tool, and it's priced for that intensity.
Weighing an AI rollout? Talk it through with a Claude partner →
What do Team and Enterprise add for organizations?
Individual plans belong to individuals — which becomes a problem the moment a company wants shared billing, central administration, and control over where its data goes. That's what the organizational tiers solve.
- Team gives a company centralized billing and admin over a group of seats, plus collaboration features — the sensible floor for any business rolling Claude out beyond a couple of enthusiasts.
- Enterprise layers on what larger organizations require before legal signs anything: enhanced security and compliance controls, identity integration, and administrative depth for managing Claude across departments.
One pattern worth flagging: companies often start as a handful of individual Pro accounts on personal cards. It works until it doesn't — no admin visibility, no unified data agreement, no offboarding control. Moving to Team early is cheaper than untangling shadow accounts later.
What drives Claude costs up?
Three levers, whichever side of the lineup you're on:
- Seats. On subscription plans, cost scales with headcount — straightforward, predictable, and worth auditing occasionally for seats nobody uses.
- Usage volume. On the API, tokens are the meter. Long documents, long conversations, and high request volume all turn it faster. This is the number to model carefully before launching anything customer-facing.
- Model choice. The most capable model isn't always the right one. Plenty of production workloads run beautifully on faster, cheaper models — and routing easy tasks to light models while reserving heavyweights for hard ones is the single biggest API cost optimization we implement.
Before committing budget, it's worth running the value side of the equation too — our guide to calculating AI ROI before you buy a license walks through the math.
Which Claude plan should your business start with?
Our honest defaults: individuals start Free and upgrade to Pro when they hit limits — the limits themselves will tell you. Businesses adopting Claude as a work tool should start with Team for the admin and data controls alone, moving to Enterprise when security review or scale demands it. Anyone building Claude into a product or internal workflow prices out the API with a small pilot before projecting costs at volume — token math on paper and token math in production are rarely the same.
For current pricing, go straight to anthropic.com — always the source of truth for numbers. And if you want help choosing, licensing, and actually deploying Claude — not just buying seats that sit idle — that's what we do as an Anthropic partner. Start with our Claude licensing overview or book a scoping call and we'll map tiers to your team, workflows, and volume.
Frequently Asked Questions
What are Claude's subscription plans?
As of this writing, Claude offers Free, Pro, and Max plans for individuals, and Team and Enterprise plans for organizations, all priced per seat. Developers building on Claude use the API instead, which bills by usage in tokens. Current pricing is published at anthropic.com.
How much does Claude cost?
It depends on which tier and how it's billed: subscription plans are flat per-seat rates that rise with tier, while API costs scale with token usage and model choice. Anthropic adjusts pricing over time, so check anthropic.com for current figures rather than trusting blog posts.
Should my business buy Claude Team or just individual Pro seats?
Team, in almost every case. Individual Pro accounts on personal cards leave a business with no admin visibility, no centralized billing, and no offboarding control. Team adds central administration and collaboration features, and Enterprise extends that with security, compliance, and identity integration for larger organizations.