Claude AI Plans Compared: Free, Pro, Max, Team, Enterprise

Claude sells three kinds of plan. Individuals choose Free, Pro ($17-20), Max ($100) or Ultra ($200). Organizations choose Team ($20 or $100 per seat, 2 to 150 people) or Enterprise ($20 per seat plus usage at API rates). Universities get an Education plan, and developers pay per token. Prices checked 2026-10-31.

Prices checked 2026-10-31

All figures come from claude.com/pricing and Anthropic's help center. Tax is excluded, and Anthropic reserves the right to change plans and prices.

This page is the map. Each plan has its own page when you want the depth: Claude pricing for the main overview, Claude Pro pricing, Claude Max plan, Claude API pricing and Claude Code pricing.

The ladder, in one view

Think of it as four questions, asked in order.

1. Do you need Claude Code, Opus or Research? If not, Free is enough. If so, Pro is the floor.
2. Do you hit limits weekly? If so, Max or Ultra.
3. Is more than one person involved? If so, Team or Enterprise.
4. Is software calling Claude? If so, the API, whatever plan you hold.

Feature matrix, individual plans

Context window is listed as "up to 1M, varies by model" on every individual plan. Claude Motion (beta) and Claude Tag are marked unavailable on all individual plans, and sharing and collaboration on Projects is a team feature.

Feature matrix, organization plans

Where the Team plan offers Fable, it is for Premium seats only. Standard seats get Fable through usage credits.

Team: how it is built

Team is for 2 to 150 people. The minimum is two members. There are two seat types, and you can mix them.

The API credits from all seats pool into one balance, capped at $500 a month. Anthropic's example: three Standard and two Premium seats earns $260 a month. Adding a Premium seat raises the next cycle's credits to $360. One Owner links a Console organization to claim it.

Team adds a layer over Pro: central billing and administration, SSO with domain verification and just-in-time provisioning, role-based permissions, spend caps per user and organization, enterprise search across Slack, Microsoft 365 and custom connectors, and connectors for Google Drive, Gmail, Calendar and GitHub. Anthropic's Team article lists a 200K context window for Team; the pricing page says up to 1M varies by model.

Enterprise: what changes

Enterprise is priced at $20 per seat per month, billed annually, with a 20-seat minimum, plus usage billed at API rates. Anthropic says cost scales with the models and tasks the team runs, so the seat fee is only the floor. Usage-based Enterprise bills consumption instead of a seat allowance.

What Enterprise adds over Team is governance: role-based access with fine-grained permissions, SCIM, audit logs, a Compliance API, custom data retention, network-level access controls, IP allowlisting, inference hooks, a HIPAA-ready offering and admin-set user and organization spend limits.

There are two ways to buy. Self-serve Enterprise is a credit-card or ACH purchase. Sales-assisted adds invoicing and net terms, AWS Marketplace availability, consultation, tiered incentives on committed spend, trials, customer-success support at spend thresholds and Claude Security (beta).

Education

Anthropic offers a university-wide Education plan covering students, faculty and staff at discounted rates, with academic research and learning mode, dedicated API credits and training resources. No public price is listed. Eligible nonprofits can get discounted Team and Enterprise pricing instead. Individuals do not get student discounts on Pro, Max or Ultra.

How usage limits compare

Limits are relative, not absolute. Everything is measured against Pro.

Every limit is shared across chat, Claude Code and Cowork. When someone reaches a limit, there are three outs: wait, move up a plan, or turn on usage credits, which continue work at standard API rates until a spend cap you set. Team and Enterprise admins set those caps at organization, group or user level.

What changes the answer

Data. Individual plans are opt-out for model training. Team and Enterprise default to none. If you are putting customer or client work through Claude, that matters more than any price gap.

Admin. If you need to know who is spending what, Team has analytics and spend controls. If you need an audit trail, you are on Enterprise.

Model access. Fable follows the premium seat rule everywhere: included on Max, Ultra and Team Premium, credits everywhere else.

Billing. Max and Ultra are monthly only. Annual is available on Pro and Team, and Enterprise is annual only. Payments are generally non-refundable.

Who should pick which

Our take

Claude's plan structure is clean: three individual steps that scale usage 1x, 5x and 10x, then two organization products that add admin. The part that needs care is how they interact. Usage is one pool across chat, Claude Code and Cowork, which makes it easy to buy the wrong plan if you only think about chat.

The most useful check is the seat arithmetic. A Team Standard seat at $20 annual matches Pro, so a two-person company pays the same per head and gains admin and sharing. A Premium seat at $100 matches Max and carries the pooled API credit. Past about 20 seats, Enterprise adds the controls auditors ask for, at the same $20 seat price plus API-rate usage.

The things we cannot tell you from the published pages are how many tasks a plan holds, and how a Team Standard seat's usage compares with Pro in numbers; Anthropic only says "more". If that number decides your purchase, ask Anthropic or run a pilot. For model choice inside a plan, see Opus vs Sonnet and Claude models. For competing seat pricing, see Cursor pricing and GitHub Copilot pricing.

Checklist

• List who needs Claude Code, who needs only chat, and who needs admin.
• Decide on training defaults before putting client data in.
• Price Premium seats only for named heavy users.
• If you want the monthly API credit, nominate the Owner who links the Console organization.
• Check Enterprise's 20-seat minimum before planning a pilot.
• Revisit plan choice after a month of usage data from the admin console.

Questions about moving between plans

Can I change plans mid-month? Yes, with rules that differ by direction. On individual plans an upgrade applies the new limits right away. Moving from annual Pro to Max or Ultra gives an account credit for the unused balance if it exceeds the new price, provided the billing address matches. Max to Ultra is prorated, and API credits you already claimed are prorated too. Switching a monthly Pro to annual credits unused time toward the first annual invoice.

What happens to API credits if I move to Team? The Max or Ultra link ends. A Team Owner or Primary Owner can claim the team's credits once the Team plan has been active for seven days, and the pool is calculated from the seats on the plan each time credits are deposited.

Do Team and Enterprise members keep their personal plan? Anthropic's pages we read do not say.

How do seats change mid-term? The pricing page lists adding seats mid-term as available on Team and both Enterprise routes.

What does the Education plan cost? Not published; the page routes you to a sales conversation.

A note on how to budget a team

Start with the seat count and a seat type for each person, then add a usage-credit cap at organization level instead of buying Premium seats for everyone. Standard seats cost a fifth of Premium seats, and the admin console shows estimated spend per user once usage credits are on, so you can see who actually hits limits and move only those people up. Claude Code in particular has a spend report with CSV export, updated daily, and an analytics dashboard of daily active users and sessions. A month of that data is a better basis for choosing Premium seats than a guess.

If you are an agency or a company with client data, read the data row of the matrix before the price row. Team and Enterprise default to no training on your content, while individual plans are opt-out. That difference can decide the purchase on its own, and it costs $20 a seat to fix on Team.

Reading the matrix as a buyer

Three rows decide most purchases. The first is Claude Code, which is absent only on Free. The second is Fable, which is the one model whose access changes by seat tier rather than by plan name. The third is the security block: single sign-on, audit logs and SCIM appear only as you move from Team to Enterprise. If none of those three matter to you, the cheapest plan that clears your usage need is the right one, and the matrix has little else to say. Re-run the exercise each quarter, because Anthropic has moved features between tiers and renamed plans within the year.