Skip to content
Claude Code pricing guide

Claude Code pricing: what it costs on a plan, the API and a team.

Claude Code is included in every paid Claude plan, not Free: Pro at $20 if billed monthly, Max at $100 per month and Ultra at $200 per month. With a Claude Console account you pay per token. These are Anthropic’s US list prices as of 14 November 2026, before tax, and a plan’s checkout may use a local currency.

By Last updated

Claude Code pricing in short

  • Claude Code is included in all paid Claude plans. The Free plan does not include it. Pro, Max, Ultra and Team seats come with a usage allowance; an Enterprise seat gives access only, with all usage billed on top.
  • Individual plans: Pro is $20 if billed monthly, or $200 billed up front for a year, which the pricing page shows as $17 per month. Max is $100 per month. Ultra is $200 per month.
  • Team: a Standard seat is $20 per seat per month if billed annually, or $25 if billed monthly. A Premium seat is $100 per seat per month if billed annually, or $125 if billed monthly.
  • Enterprise: US$20 per seat per month, billed annually, with a 20 seat minimum, plus usage at API rates.
  • Through the Claude API, with a Claude Console account, you pay per token, in US dollars. Sonnet 5.5 is $2 per million input tokens and $10 per million output tokens; Opus 5.5 is $4 and $20.
  • These are US list prices on Anthropic’s pages, in US dollars, as of 14 November 2026. They don’t include applicable tax, and the checkout for a plan may use a local currency where Anthropic supports one. Check the final total and any tax at checkout.

Sources, read 14 November 2026: Claude plans and pricing, Set up Claude Code, What is the Team plan?, Claude API pricing.

Which Claude plans include Claude Code, and what do they cost?

Every paid plan does. Anthropic’s pricing page says “Claude Code is included in all paid plans”, and its setup page says “The free claude.ai plan does not include Claude Code access.” So for one person the question is which paid plan. These are the US list prices on Anthropic’s pricing page as of 14 November 2026, which says: “Prices shown don’t include applicable tax.”

PlanPrice (US list, before tax)Claude CodeBilling
Free$0Not includedNo charge
Pro$20 if billed monthly, or $17 per month with the annual discount ($200 billed up front)IncludedMonthly or annual
Max$100 per monthIncludedMonthly only
Ultra$200 per monthIncludedMonthly only

Max and Ultra give more usage than Pro: Anthropic’s articles What is the Max plan? and What is the Ultra plan? have the detail.

What the plan price buys

Pro, Max, Ultra and Team seats include a usage allowance shared across Claude and Claude Code. Inside Claude Code, /usage shows it as a bar, not a bill:

Open the full-size screenshot: Claude Code’s /usage screen on its Usage tab: session cost and token counts at zero, then Current week (all models) with a bar at 10% used and Resets Nov 20 at 2pm (Europe/London)
/usage in Claude Code 2.1.323 on an Ultra plan (13 Nov 2026): the plan’s week across all models, 10% used, and when it resets.

The limits, when they reset and every way to check them are in our Claude usage guide.

Other currencies, tax and mobile prices

The dollar figures above are what Anthropic’s pricing page shows, not a promise of your total. Its Pro page calls $20 the US price, “with pricing in your local currency where supported”, and says “some regions include applicable taxes in the displayed price while others add tax at checkout”. Its invoice page for Pro, Max and Ultra says “Tax is calculated from your billing address”, and that the currency is the one offered to new subscribers in your region. The Max and Ultra pages add: “This price is for web subscriptions only. Mobile pricing may vary depending on your app platform.” Check the final total and any tax at checkout before you confirm.

What a plan can cost beyond its price

Usage credits are billed “at standard API rates”, and fast mode and the Fable models on Pro plans and Team Standard seats use them before the plan’s ordinary limits are reached: Anthropic’s article Manage usage credits for paid Claude plans has the rest.

Sources, read 14 November 2026: Claude plans and pricing, Set up Claude Code, Use Claude Code with your Pro, Max, or Ultra plan, What is the Pro plan?, What is the Max plan?, What is the Ultra plan?, Understanding your Pro, Max, or Ultra plan invoices, Manage usage credits for paid Claude plans, Claude Fable models on your plan, Fast mode in Claude Code.

What does Claude Code cost through the Claude API?

With a Claude Console account, Claude Code is billed per token at Claude API prices, separately from any Claude plan. Most Console organizations pay with prepaid usage credits, bought before use; Anthropic says purchased credits “cover API access, playground usage, and Claude Code”. Anthropic says all API prices and payments are in US dollars. These are the four models its pricing page lists under “Latest models” as of 14 November 2026, per million tokens. Its API pricing page lists the older models too.

ModelInputCache write (5 minutes)Cache readOutput
Fable 5.2$8$10$0.20$40
Opus 5.5$4$5$0.20$20
Sonnet 5.5$2$2.50$0.10$10
Haiku 5.5, prompts up to 100,000 tokens$0.10$0.125$0.01$0.50
Haiku 5.5, prompts over 100,000 tokens$0.50$0.625$0.05$2.50

Each figure is the price per million tokens. Input is what Claude Code sends the model: your prompt, files, the conversation so far and tool results. A cache write stores part of that so the next request can reuse it, and a cache read is the reuse, at a fraction of the input price. Output is what the model writes back.

What raises the API price

  • Fast mode. Anthropic’s API pricing page lists fast mode at $16 per million input tokens and $80 per million output tokens for Fable 5.2, and $8 and $40 for Opus 5.5: twice the standard price.
  • US-only inference. For Claude 4.6 and later models, asking for US-only inference multiplies input, output, cache write and cache read prices by 1.1 on the Claude API, on Claude Platform on AWS and on Claude Models on Google Cloud where Anthropic manages them. On Claude in Microsoft Foundry, the same 1.1x multiplier applies to deployments that use the US Data Zone Standard deployment type. Amazon Bedrock and the Google-managed Claude Models on Google Cloud have their own regional pricing.
  • One-hour cache writes. A write to the one-hour cache costs 2 times the base input price, against 1.25 times for the five-minute cache in the table.
  • Long prompts on Haiku 5.5. A request whose prompt is over 100,000 tokens pays the higher row in the table.

For scale, Anthropic’s Claude Code docs give one benchmark, for enterprise deployments and not a forecast of anyone’s bill: “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”.

A worked example

This sum is ours, with made-up token counts, and it leaves out any cache-write charge. One request on Sonnet 5.5 sends 50,000 fresh input tokens, reads 200,000 from the cache and gets 10,000 tokens back. That is 0.05 × $2, plus 0.2 × $0.10, plus 0.01 × $10: $0.10 + $0.02 + $0.10, or $0.22. The same counts come to $0.44 on Opus 5.5 and $0.84 on Fable 5.2. A session is many requests.

Claude Code shows a running estimate of a session’s cost, and our Claude Code status line guide shows how to keep it on screen.

An API key can override your plan

In an interactive session, if an ANTHROPIC_API_KEY environment variable is set, Claude Code asks once whether to use it and remembers your choice. If you approve it, Claude Code uses that key instead of your subscription, “resulting in API usage charges rather than using your subscription’s included usage”. In non-interactive mode (-p) “the key is always used when present”. Cloud sessions “always use your subscription credentials”. Anthropic’s advice for subscribers is to keep the variable unset and to run /status to see which account Claude Code is using.

Through a cloud provider

Claude Code also runs on Amazon Bedrock, Google Cloud and Microsoft Foundry, for one person or for an organization. There it “is billed per token to your cloud account”, and the price can differ from the table above: for Amazon Bedrock and the Claude models Google manages, Anthropic’s API pricing page points to Amazon’s and Google’s own price lists. Anthropic’s other routes for an organization are Team seats, Enterprise seats and the Claude Console.

Sources, read 14 November 2026: Claude API pricing, Claude plans and pricing, Manage costs effectively, How do I pay for my Claude API usage?, Manage API key environment variables in Claude Code, Authentication.

What do Claude Code team seats cost?

Claude Code is included with every Team plan seat and with the Enterprise seat. The two plans charge in different ways: a Team seat is a flat price with its own usage limits, and an Enterprise seat is a fee for access with all usage billed on top. These are the US list prices on Anthropic’s pricing page as of 14 November 2026.

SeatPrice (US list, before tax)UsageSize and billing
Team, Standard seat$20 per seat per month if billed annually, or $25 per seat per month if billed monthlyIncluded, with more usage than Pro2 to 150 seats. Monthly or annual
Team, Premium seat$100 per seat per month if billed annually, or $125 per seat per month if billed monthlyIncluded, with 5x more usage than Standard seatsStandard and Premium seats can be mixed in one team
Enterprise seatUS$20 per seat per month, billed annuallyNot included: every token is billed at standard API rates20 seat minimum on a self-serve plan, 50 through Anthropic’s sales team. Annual only

Team seats

A Team plan needs at least two members and supports up to 150 seats. Each seat is for one named person, and the usage limits are per member: one person reaching a limit does not affect anyone else. The organization can turn on usage credits so members keep working past their limits. Anthropic’s Team page says its prices “are for US customers and exclude applicable taxes”, and that “Pricing, currency, and tax handling vary by region”.

Enterprise seats

On the current Enterprise plan the seat fee “doesn’t include any usage. Every token your team consumes is billed separately at standard API rates”. So the real cost is the seat fee plus what the team uses, at the per-token prices in the table above. On a self-serve plan that usage is bought up front as credits, in US dollars only. On a sales-assisted plan it is billed monthly in arrears. Admins can set spend limits for the organization and for each user. Some older Enterprise contracts still have seat types that include a usage allowance; Anthropic says they move to the single seat at renewal. Anthropic’s pages differ on this: its Claude Code costs page still describes Enterprise usage as drawing from a seat allowance. This guide follows the pricing page and the Enterprise billing articles. Its pricing page adds that Enterprise and API pricing can be customized with volume and committed-spend discounts through its sales team, and that it does not publish discount rates.

A team on the Claude Console

Anthropic’s docs list the Claude Console as a separate route, for “organizations that prefer API-based billing”: an admin invites developers to a Console organization, and their Claude Code usage is billed per token to it, with spend limits set per workspace.

Sources, read 14 November 2026: Claude plans and pricing, What is the Team plan?, Use Claude Code with your Team or Enterprise plan, What is the Enterprise plan?, How am I billed for my Enterprise plan?, Authentication, Manage costs effectively.

Questions people ask

Is Claude Code free?

Not through the Free plan: Anthropic says “The free claude.ai plan does not include Claude Code access.” Claude Code needs a Pro, Max, Ultra, Team, Enterprise or Console account, or a cloud provider such as Amazon Bedrock. Its API pricing page says “New users receive free credits to test the API”, without giving an amount.

Is Claude Code included in Claude Pro?

Yes. Pro is the lowest-priced individual plan that includes it: $20 if billed monthly, or $200 billed up front for a year, as of 14 November 2026. Those are US list prices before tax.

Does Claude Code cost extra on top of a Claude plan?

On Pro, Max, Ultra and Team, not for ordinary usage inside the plan’s limits. Usage credits are billed separately at API rates, and so is anything Claude Code does on an API key or a Console account. On the current Enterprise plan every token is billed on top of the seat fee.

Can I pay for Claude Code yearly?

On Pro and Team, yes: Anthropic offers annual subscriptions for those two plans. Max and Ultra are monthly subscriptions only, and Enterprise is annual only.

Do the monthly API credits on Max, Ultra and Team pay for Claude Code?

No. Anthropic says the monthly API credits “can’t be used in Claude Code or the Claude apps”: see Monthly API credits for Max, Ultra, and Team plans.

Is a plan or the API cheaper for Claude Code?

Anthropic gives no rule, and it depends on how much you use. Its pages give the sides: Pro, Max, Ultra and Team seats are a fixed price with usage limits, the API is a price per token, billed for what you use, and an Enterprise seat is a seat fee plus per-token usage. Its pricing page says that for heavy coding sessions “you can also switch to pay-as-you-go API credits through a Console account”.

How do I see what Claude Code has cost me?

On a plan, our Claude usage guide shows every place your limits appear. On the API, the Claude Console’s Cost page shows the spend.

Sources, read 14 November 2026: Set up Claude Code, Claude plans and pricing, Claude API pricing, Monthly API credits for Max, Ultra, and Team plans, Manage costs effectively.