Claude Team vs Enterprise pricing (2026): per-seat costs and what each tier includes

Anthropic sells Claude to organizations through two business tiers: Team, a self-serve per-seat plan with a 2-seat minimum, and Enterprise, a sales-led annual plan at $20 per seat plus usage at API rates. Team comes in two seat types (Standard at $20/seat/mo annual and Premium at $100/seat/mo annual), and both bundle Claude Code on every seat. Enterprise is where SSO, SCIM provisioning, audit logs, and a Compliance API live. This guide gives you the per-seat numbers, a real Team-vs-Enterprise feature matrix, and a plain decision on which tier is and is not worth it for your rollout. Every price here is sourced from claude.com/pricing and the Anthropic docs pricing reference, verified on 2026-09-25.

What does every Claude plan cost, from Free to Enterprise?

Every Claude plan is one of five tiers: Free, Pro, Max, Team, or Enterprise, each priced by usage allowance and admin features rather than by a fixed message count. The table below is the at-a-glance map; the rest of the guide drills into the two business tiers. Every figure is verified against claude.com/pricing on 2026-09-25, but Anthropic reprices seats and limits often, so confirm before you buy.

The two consumer plans (Pro and Max) and the API are covered in depth in our companion piece, every Claude plan compared. This page is the rollout buyer's view: per-seat economics, minimums, and what Enterprise actually adds.

What is the difference between a Team Standard seat and a Premium seat?

A Team seat is a per-user license on Anthropic's self-serve business plan, and it comes in two usage tiers that you can mix on the same team. Standard is the everyday seat; Premium is the heavy-usage seat for people who lean hard on Claude Code or very long context.

Two numbers are load-bearing here. First, the minimum is 2 seats (the maximum is 150), so the floor for a Team account is 2 Standard seats at $20 each ($40/mo annual billingverified 2026-09-25), or 2 Premium seats at $100 each ($200/mo annual). Second, annual billing saves roughly 20 percent versus monthly: $20 instead of $25 on Standard, $100 instead of $125 on Premium. These rates come straight from claude.com/pricing.

What do both Team seats include?

Both Team seat types bundle the same core capability set, differing only in how much usage you get per seat. According to the Anthropic Team plan support article, every seat includes:

• Claude Code on every seat, the agentic coding tool, under one unified subscription with no separate API key needed
• More usage per seat than Pro, so each team member gets a higher allowance than a $20 individual Pro plan
• Central admin controls, seat management, usage visibility, and managed settings
• Centralized billing, one invoice for the whole org instead of per-person card charges
• Shared workspaces, so teammates can collaborate on projects and prompts

The thing to internalize: Claude Code being seat-inclusive is the differentiator. On a per-seat Team plan, Claude Code usage draws from the plan's shared session and weekly limits rather than being metered per token, so a developer is not also paying a separate API bill to use it. For how that compares to running Claude Code on a raw API key, see our plan-by-plan breakdown.

The thing to internalize: Claude Code being seat-inclusive is the differentiator.What both Team seats include

Is the Premium seat math worth it over Standard?

The Premium seat is a $80/seat/month upgrade over Standard whose entire value is more usage for heavy Claude Code and long-context users. The honest framing: a Premium seat at $100/mo annual costs the same as an individual Max 5x plan, so you are paying Max-tier money but getting it inside the team's admin and billing envelope.

The break-even logic is usage frequency, not headcount. A Standard seat gives more usage than Pro, which is plenty for someone who uses Claude for a few dozen chats and the occasional Claude Code session per day. The moment a team member starts hitting the Standard seat's session limit several times a week, the Premium seat (5x a Standard seat's usage, per claude.com/pricing) pays for itself in unblocked work. If nobody on the team is hitting Standard limits, Premium is wasted money.

How does Claude Enterprise bill?

Claude Enterprise is bought through sales and billed annually, and its structure is published: claude.com/pricing lists it as $20 per seat per month plus usage billed at API rates, so the total scales with the models and tasks your team runs. What gets negotiated is volume discounts, custom limits, and support, not the shape of the bill.

• The seat fee, $20 per seat per month on an annual contract: the fixed part of the bill, and the part a headcount forecast predicts.
• Usage on top, billed at standard API rates measured in tokens (for example, Claude Opus 5.5 at $4 input / $20 output per million tokens, and Fable 5.1, Anthropic's top model, at $10 / $50, per Anthropic's pricing docs on 2026-09-25). This is the part that varies: automation-heavy teams pay far more than light chat users.

For volume buyers, the negotiation levers are custom rate limits, volume discounts, and dedicated support, reached through claude.com/contact-sales or [email protected]. There is no coupon or self-serve discount code; the only real price lever short of Enterprise is annual-versus-monthly billing on Team.

What does Enterprise add over Team (SSO, SCIM, audit logs)?

Enterprise-only features are the identity, security, and compliance controls that a self-serve Team plan does not carry. These are the reasons a regulated or large org has to move off Team:

• SSO / SAML, single sign-on against your identity provider (Okta, Entra ID, and similar)
• SCIM provisioning, automated user provisioning and deprovisioning so leavers lose access automatically
• Audit logs, a record of who used Claude, when, and from where, for security review
• Compliance API, programmatic export of usage and content data for eDiscovery and governance
• HIPAA support, available under Enterprise terms for healthcare workloads
• Custom data retention, configure or zero out how long conversation data is retained
• Larger context and custom rate limits, negotiated rather than fixed
• Dedicated support, a named contact and priority channel

If none of those words appear in your security team's requirements, you almost certainly do not need Enterprise yet, and Team will serve you at a fraction of the procurement friction.

Team vs Enterprise: the full feature and seat matrix

This is the comparison that decides the rollout. The matrix below contrasts the two business tiers across the dimensions a buyer actually weighs: per-seat cost, minimum seats, identity, admin, and compliance. Figures verified 2026-09-25 against claude.com/pricing.

Read it as a single line: Team is the self-serve, per-seat default for any group of 2 to 150; Enterprise is what you escalate to the moment SSO, SCIM, audit logs, or HIPAA appear on the requirements list.

Should one heavy user buy Max or a Team Premium seat?

This is the most common edge case, because a Max plan and a Team Premium seat cost almost the same per person. The decision is individual-versus-organization, not price.

If you are a solo developer or a founder buying for yourself, Max wins: same usage, no 2-seat minimum, no admin overhead. The Max question is its own decision; we walk it fully in our Max worth-it analysis. If you are buying for a team and want one bill, shared context, and the ability to add or remove people centrally, the Premium seat earns its place even though the per-head price matches Max.

What admin controls do Team and Enterprise give you?

Admin controls are the org-level levers, spend, visibility, and policy, that turn a pile of individual subscriptions into a managed deployment. On Team you get the core set: add and remove seats, see usage across the org, and manage shared settings. Enterprise extends this with managed policies, deeper analytics, and the identity controls above.

For a rollout buyer, the practical admin wins are three: one invoice instead of expense-report sprawl, the ability to deprovision a leaver's seat (automated via SCIM on Enterprise), and usage visibility so you can spot who actually needs a Premium seat versus who is fine on Standard. That last one is how you keep the bill honest, start most people on Standard and promote to Premium only when the usage data says they are hitting limits.

How do usage limits and extra usage work across the business tiers?

Usage limits are the real currency of Claude pricing, because Anthropic prices by allowance windows rather than a published message count. The structure, per Anthropic's plan support docs, is a session-based limit that resets every 5 hours, plus a separate weekly limit that resets 7 days after a session starts. Higher tiers raise the ceiling; they do not remove it. Enterprise works differently: usage is billed at API rates on top of the seat fee.

When a seat exhausts its allowance, the team can opt in to extra usage credits that let users continue at standard API rates, but this requires explicit consent so an org never silently runs up an API bill. This is the one place where the otherwise flat per-seat model becomes metered, and it is worth flagging to finance before a heavy Claude Code rollout. There is no published "Pro = 45 messages, Max = 900 messages" figure that Anthropic stands behind; those numbers circulate secondhand and should be treated as illustrative, not primary-sourced.

When should you upgrade, and to which tier?

The upgrade decision is a function of headcount and compliance, not of how impressive a tier sounds. Here is the framework we recommend, in order:

1. Just you: stay on individual Pro or Max. Team needs at least 2 seats, and annual Pro ($17/mo) is cheaper than a $20 Standard seat.
2. 2 to 150 people, want one bill and admin: Team Standard. Promote individuals to Premium seats only when their usage data shows repeated limit hits.
3. Need SSO, SCIM, audit logs, or HIPAA: Enterprise. There is no way to bolt these onto Team, so a single hard compliance requirement forces the jump.
4. Spiky, automation-heavy usage: Enterprise, where usage is billed at API rates on top of the $20 seat fee, so heavy days are paid for rather than capped.

What is the migration path from Pro or Max to Team to Enterprise?

The migration path is a deliberate ladder: individuals validate on Pro or Max, the org consolidates onto Team for billing and admin, and only compliance forces the final step to Enterprise. In practice:

1. Pro / Max (individual): a few people prove the workflow on their own subscriptions. Cheap, instant, no commitment.
2. Team (2 to 150 seats): roll everyone onto Standard seats for one invoice and central control; sprinkle Premium seats on the heavy users. Self-serve, same-day.
3. Enterprise: trigger the sales cycle when security mandates SSO, SCIM, audit logs, a Compliance API, or HIPAA, or when paying API rates for usage beats hitting per-seat caps at your volume.

Most organizations live happily on Team and never need Enterprise. The jump is driven by your security and procurement teams, not by Claude's product surface, which is identical in capability across the tiers.

A worked cost savings example for a 10-person team

A worked example makes the per-seat math concrete. Take a 10-person team where 8 people do normal work and 2 are heavy Claude Code users. The cheapest correct configuration is a mix, not a single tier:

The mixed config at $360/month annual gives the 2 heavy users their Premium allowance while keeping the other 8 on cheap Standard seats, versus $1,000 if you put everyone on Premium. That $640/month gap is the entire reason to mix seats rather than pick one tier for the whole team. Billed monthly instead of annual, the same mix is (8 × $25) + (2 × $125) = $450/month, so annual billing saves $90/month here. Confirm the live per-seat rates at claude.com/pricing before you commit, since Anthropic adjusts seat pricing periodically.

That $640/month gap is the entire reason to mix seats rather than pick one tier for the whole team.10-person cost example

If your org is set up as a business, these seats are a deductible software expense; our sister guide on self-employed and small-business software deductions covers how subscription tooling reduces the effective per-seat cost.

The bottom line on Claude Team vs Enterprise

For nearly every organization of 2 to 150 people, Team is the right answer: Standard seats at $20/seat/mo annual, Premium seats at $100 for the heavy users, Claude Code on every seat, and self-serve admin. Enterprise is not an upgrade you buy for prestige; it is the tier you are forced into the moment SSO, SCIM, audit logs, a Compliance API, or HIPAA becomes a hard requirement, or when paying API rates for usage beats per-seat caps for spiky workloads. Mix Standard and Premium seats to avoid overpaying, bill annually to save about 20 percent, and reconfirm the numbers at claude.com/pricing before you commit, because Anthropic reprices this surface more often than most vendors.

Standards and research behind this page

Claims about what an AI tool can and cannot do are easiest to check against the standards bodies and research groups that measure these systems rather than against any vendor's own description. The sources below are the primary ones, linked so a reader can verify a claim here without taking this page's word for it.

• NIST AI Risk Management Framework is the US federal reference for evaluating and governing AI systems, including how their capabilities and limits should be characterised.
• NIST AI RMF Knowledge Base sets out the measurable characteristics of a trustworthy AI system, which is the vocabulary these comparisons use.
• Stanford HAI AI Index publishes the annual measured benchmarks on model capability, cost and adoption that vendor claims are checked against.
• National Science Foundation AI programme funds and documents the underlying research these products are built on.
• arXiv cs.AI carries the primary papers behind most model capability claims, usually months before a product mentions them.
• FTC Endorsement Guides governs how a review or recommendation must disclose a material connection, which is why the disclosure on this page exists.
• US Copyright Office states what copyright covers, which decides who owns the output of a generative tool.
• Section 508 defines the federal accessibility requirements a tool must meet to be usable in public-sector work.

This page compares tools and does not endorse one. Capabilities and pricing change frequently, and a reader should confirm current behaviour with the vendor before relying on it.