Claude Code for Teams: Team Seats vs API vs Enterprise (2026)

For a team of 2 to 150 developers who use Claude Code every day, a Claude Team plan is likely the cheaper way to pay, as long as their work fits inside the seat allowance: a Premium seat is $125 a month billed monthly ($100 annually), while Anthropic's own docs put the average enterprise developer at $150-250 a month at API rates. Anthropic does not publish that allowance, so treat the saving as something to confirm with a short pilot, not a given. Pick Enterprise when you need seats with SCIM, session transcripts or per-group spend caps, and keep a Console API org for CI and automation, where seats do not fit.

Last updated: October 9, 2026. Every price, limit and setting below was checked on this date against claude.com/pricing, the Claude Code costs docs and the Claude Help Center articles linked in each section.

There are four ways to put Claude Code in front of a team, and they bill in completely different units. Team seats bill a flat fee with a usage allowance. Enterprise bills a seat fee plus every token at API rates. The Claude Console bills tokens only. Amazon Bedrock, Google Cloud's Agent Platform and Microsoft Foundry bill tokens to your cloud account. Choosing wrong does not break anything, it just means paying for the wrong thing, or losing the controls your security team will ask for in month three.

At a Glance#

The Team and Enterprise prices are from claude.com/pricing. The usage-model and reporting rows are from the "Manage costs for your organization" table in the Claude Code costs docs. The SCIM row is from the Help Center's JIT and SCIM provisioning guide, which also lists the roles each product supports. The Console's per-workspace roles (Admin, Developer, Limited Developer, User) are in the Workspaces docs. The audit log and transcript rows are from the Help Center audit log article ("available for Enterprise organizations only") and the Compliance API docs, which say a standalone Console organization "can query the Activity Feed only." Anthropic's Claude Platform Compliance API announcement adds that the Console feed covers admin and resource events, not model inference, and that your account team switches it on.

How each option actually bills#

Team plan. Claude Code is included in every Team seat, per the Help Center article on using Claude Code with a Team or Enterprise plan. The costs docs describe the meter: each member's usage "draws from a per-seat allowance that resets on a rolling five-hour window and a weekly window," and that allowance is shared with Claude chat and Cowork. Anthropic does not publish the allowance in dollars. The pricing FAQ says only that Standard seats give more usage than Pro and Premium seats give 5x more than Standard. When a member runs out, an admin can turn on usage credits so they keep working "at standard API rates." You can mix seat types on one plan, so the heavy Claude Code users get Premium and everyone else gets Standard.

Enterprise. New Enterprise plans use a single seat that includes Claude Code, Cowork and chat. The Help Center's Enterprise billing article is blunt about it: the seat fee "doesn't include any usage," and "every token your team consumes is billed separately at standard API rates." Self-serve Enterprise buys usage upfront in credits, and when the shared balance hits zero, usage stops for the whole organization until an Owner buys more. Sales-assisted Enterprise bills usage monthly in arrears. Seats cannot be removed mid-term on self-serve Enterprise. The old Enterprise Standard and Premium seats are no longer sold on new contracts.

Console API. Developers sign in with a Console account, and Claude Code creates a workspace called "Claude Code" for all of that usage. You cap it with workspace spend limits and read per-member spend from the Console dashboard or the Claude Code Analytics API with an Admin API key. Claude Code traffic counts toward your organization's API rate limits, so the costs docs recommend setting a workspace rate limit to protect production workloads.

Cloud providers. On Bedrock, Google Cloud's Agent Platform and Foundry, the costs docs say Claude Code "does not send metrics from your cloud back to Anthropic," so none of Anthropic's analytics dashboards cover that usage. Per-user attribution comes from OpenTelemetry export, a self-hosted Claude apps gateway, or an LLM gateway such as LiteLLM.

The 10-developer math#

The anchor number is Anthropic's own telemetry from the costs docs: "Across enterprise deployments, the average cost is around $13 per developer per active day and $150-250 per developer per month, with costs remaining below $30 per active day for 90% of users." Those are API-rate figures, so they apply directly to the Console and to Enterprise usage. The rows below are our arithmetic on the published prices, for ten developers who all use Claude Code daily.

Read the Team rows carefully. They are only cheaper if the work fits inside the seat allowance, and because Anthropic does not publish that allowance as a dollar figure, you cannot know your break-even in advance. The practical test is a two-week pilot on Premium seats with usage credits turned on and a low spend limit: if the credit spend stays near zero, seats win clearly. If your heaviest users burn through the weekly window every week, their overflow bills at API rates anyway, and the seat fee is then a premium on top of API pricing for those people.

Enterprise is never the cheapest row on usage alone. You pay API rates for every token plus $20 a seat. What you buy is the control plane, covered below.

One billing detail that changes API costs: cache lifetime#

Claude Code re-sends your whole conversation with each request and relies on prompt caching to make that affordable. The costs docs note that the cache lifetime "is an hour on a subscription and drops to five minutes once you're drawing on usage credits; on an API key or cloud provider, it's five minutes by default." A developer who steps away for twenty minutes on a Team seat comes back to a warm cache. The same developer on an API key pays to rebuild the cache on the next message. On long sessions that is a real difference, and it is one more reason interactive work tends to cost less on seats. API users can opt in. The Claude Code prompt caching docs say: "If you sign in with an API key or use a cloud provider, set promptCacheTtl to 1h to give the main conversation a one-hour cache." That needs Claude Code v2.1.242 or later. The trade-off is that one-hour cache writes are billed at a higher rate, so it only pays off for sessions that sit idle past five minutes. For how caching is priced on the API side, see our prompt caching production guide.

What Enterprise buys that Team does not#

The pricing page's comparison table lists what only Enterprise has: role-based access, SCIM, audit logs, the Compliance API, custom data retention, a HIPAA-ready offering and inference hooks. The Enterprise consumption guide adds the spend controls that matter for Claude Code specifically:

• Spend limits at three levels, organization, group and individual, with individual limits overriding group limits "regardless of which is higher."
• Pooled group budgets (beta), one shared monthly amount per group. When the pool runs out, "usage pauses for every member of the group."
• Gating Claude Code by group, so only engineering roles get the high-intensity surfaces. The guide's advice: "Giving everyone Claude Code and Cowork access on day one is the fastest way to generate unexpected consumption."
• Model and effort caps by role, enforced in Claude Code from CLI 2.1.199.

A Console org is not the governance gap it looks like next to Team: it supports SCIM when it has its own parent organization, has organization and workspace roles, and can get the Compliance API Activity Feed. What it lacks is seats, and with them the Claude Code session transcripts that only Enterprise accounts produce.

If you are at 150 people or close to it, the choice is made for you: the Team plan is for teams of 2 to 150.

Keep a Console org for CI#

Seats are tied to people who sign in, and pipelines do not run /login. Anthropic's Console to Enterprise migration guide gives two options: keep a Console org and its API keys for automation and move only interactive developers to seats, or run claude setup-token while signed in to the Enterprise organization and set the printed token as CLAUDE_CODE_OAUTH_TOKEN in CI. Most teams end up with the first: seats for people, a Console workspace with a spend limit for headless runs and scheduled agents. The guide also notes that a user can belong to both orgs at once.

The migration trap: silent billing to the old org#

The same migration guide documents the failure that will bite you when moving developers from API keys to seats. If ANTHROPIC_API_KEY, ANTHROPIC_AUTH_TOKEN or an apiKeyHelper setting is still present, it takes precedence over the new login: "/login will appear to succeed, but usage continues billing to the old Console organization." Check shell profiles, dotfiles, CI configs and the env block of ~/.claude/settings.json and repo .claude/settings*.json files.

The fix Anthropic recommends is a managed setting that pins the organization:

{
  "forceLoginOrgUUID": "your-enterprise-org-uuid",
  "forceLoginMethod": "claudeai"
}

With forceLoginOrgUUID set, Claude Code refuses to start while a leftover credential is present, which turns silent misbilling into a loud error. Run /status after signing in to confirm which organization is active. Deliver the pin as an endpoint-managed file or MDM policy, because server-managed settings only arrive after the user signs in. For the rest of the managed settings surface, see our Claude Code permissions and settings guide.

Pick by situation#

Watch for the API credit confusion#

Team plans now include a monthly Claude API credit of $20 per Standard seat and $100 per Premium seat, pooled into one balance capped at $500, per Anthropic's API credits for Max and Team plans page. That page lists Claude Code under "Doesn't cover," so the credit does not pay for Claude Code. The Help Center says plainly that "Monthly API credits included with Team plans can't be used in Claude Code or the Claude apps, including for extra usage." It is useful for the CI and Agent SDK side of the setup above, and we cover the details in Claude Max API credits: what they cover.

FAQ#

Is Claude Code included in the Claude Team plan?#

Yes. Claude Code is included with every Team seat, Standard or Premium, and with the single seat on new Enterprise plans. Usage on Team draws from the seat's allowance, which is shared with Claude chat and Cowork.

How much does Claude Code cost per developer?#

Anthropic's costs docs put the enterprise average at about $13 per developer per active day and $150-250 per developer per month at API rates, with 90% of users under $30 per active day. On a Team plan, the seat price is the cost unless a developer runs past the allowance and you have usage credits turned on.

Is the Claude Team plan cheaper than the API for Claude Code?#

It is likely to be, for developers who use Claude Code daily and whose work fits the seat allowance. A monthly-billed Premium seat is $125, below the $150-250 monthly API-rate average in Anthropic's docs. Anthropic does not publish the seat allowance in dollars, so confirm with a short pilot before you commit, and remember that overflow on usage credits bills at API rates.

What is the difference between Standard and Premium seats?#

Premium seats give 5x more usage than Standard seats, per Anthropic's pricing FAQ, and the pricing table lists Fable on Team as available on Premium seats, at up to 50% of weekly limits. Standard is $20 a month annually or $25 monthly; Premium is $100 or $125.

Can a Team or Enterprise seat be used in CI?#

Not through /login. Either keep a Console org with API keys for automation, or run claude setup-token while signed in to your Enterprise organization and set the token as CLAUDE_CODE_OAUTH_TOKEN in CI, as described in Anthropic's migration guide.

Sources#

Continue Reading#

• What a Fleet of Claude Agents Actually Costs - the same budget math once each developer runs several agents in parallel
• Claude Max API Credits: $100 a Month, What They Cover - what the Team plan's pooled API credit can and cannot pay for
• AI Coding Tools Pricing Comparison 2026 - how Claude Code's seats compare with Cursor, Codex and Copilot team plans
• The $500M Claude Bill: A Spend-Guardrails Playbook - the order to set limits in before a rollout
• Claude Code Usage Limits Playbook - what to do when a seat's 5-hour or weekly window runs out