Claude Pricing 2026: Every Plan, Usage Limit and Model Included

Claude costs $0 on Free, $20 a month on Pro ($17 a month when you pay $200 for a year), from $100 on Max and from $20 a seat on Team. Fable is included on Max, on Team Premium seats and on Enterprise. The sections below fill in limits, model access and API rates.

Short answer

• Free to try Claude before you pay. You get Sonnet and Haiku.
• Pro at $20 if you use Opus or Claude Code most days.
• Max from $100, or Team from $20 a seat, if you keep hitting Pro limits or need Fable inside the plan.
• The API if your usage is occasional or runs from scripts.

How much does each Claude plan cost?

Every price is in USD and tax is not included. Free is $0, Pro is $20 a month, and the top individual plan, Max 20x, is $200 a month. Annual billing drops Pro to $17 a month and Team seats by about 20%.

Pro's $17 means $200 up front for a year. Team takes 2 to 150 seats; Enterprise starts at 20 seats billed annually, with model usage charged at API rates on top. Anthropic's pricing page lists prices in US dollars only, before tax.

Free does not include Claude Code; Claude Code is included in all paid plans.

Which Claude plan includes Fable and Opus?

Fable is included on Max, on Team Premium seats and on Enterprise. Free has neither Fable nor Opus; Pro adds Opus 5.5, and you can run Fable with paid usage credits.

Sonnet 5.5 and Haiku 5.5 are on every plan, including Free. Fable 5.1 is the model that Fable access refers to today. Fable 5.5 has not shipped: our Fable 5.5 release-status guide tracks what Anthropic has confirmed so far.

If Fable is the reason you are upgrading, Max 5x at $100 is the cheapest plan with Fable included; for light Fable use, Pro plus paid usage credits can cost less. Team Standard seats do not list Fable, so a team that needs it has to buy Premium seats.

What do the usage limits mean?

Every plan has usage limits, and none of them is a fixed number of messages. What you get depends on conversation length, the model you pick and the features you turn on.

• Pro gives at least 5x the Free allowance per five-hour session.
• Max 5x gives 5x Pro, and Max 20x gives 20x Pro.
• Team Standard seats give more than Pro; a Team Premium seat gives 5x a Standard seat.

Web, desktop, mobile and Claude Code all draw from one pool, so a long Claude Code session spends the same allowance as a long chat. Weekly and monthly caps can apply on top of the five-hour window. On Max and Team Premium, Fable can use up to 50% of your weekly limit. The context window reaches 1M tokens and varies by model.

That shared pool is why one number cannot tell you how many messages you have left. A short question in a fresh chat is cheap; the same question at the end of a long coding session carries far more context and uses more of the allowance.

What do Max and Team add besides usage?

Max adds $100 (5x) or $200 (20x) a month in Claude Platform API credits on top of the bigger allowance. Team adds 2 to 150 seats and up to $500 a month in pooled API credits.

Claude Code is included in all paid plans, so you do not buy it separately. The API credits matter if you also call Claude from a script or an internal tool: they pay for Claude Platform usage at API rates.

Team Premium seats are the ones with Fable and 5x Standard usage. Standard seats still get Opus, Sonnet, Haiku and more usage than Pro, which is usually enough for people who mostly chat.

How much does the Claude API cost?

The API has no monthly fee; you pay per token. These are Anthropic's rates per 1M tokens, checked October 24, 2026.

Haiku 5.5 charges $0.50 per 1M input tokens once a prompt passes 100K tokens; its $0.01 cache-read rate applies up to 100K. Cache reads only hit when the start of your prompt matches what the provider stored, which is why a repeated system prompt is cheaper than a repeated question.

Take a request with 20,000 input tokens and 2,000 output tokens, no cache:

The Sonnet 5.5 line is 20,000 × $2 / 1M = $0.04 plus 2,000 × $10 / 1M = $0.02. Opus 5.5 uses $4 and $20 per 1M, so the same request costs $0.12. Fable 5.1 uses $10 and $50, so it costs $0.30. If that 20,000-token input is a cached prefix on Sonnet 5.5, the input side falls to $0.002 and the request costs about $0.022.

Plan or API: which is cheaper for you?

$20 of Pro buys about 333 Sonnet requests at $0.06 each. That is roughly eleven a day for a month. If you send more than that, a plan usually wins; if you send a few a week, or a script sends them, the API usually wins.

Claude Code has its own cost pattern because long sessions read a lot of cached context. The Claude Code pricing guide works through those numbers. If you would rather run Claude Code on an API key than a subscription, the API-key setup guide shows the two settings you need.

You can call Fable 5.1, Opus 5.5, Sonnet 5.5 and Haiku 5.5 with one OmniaKey key and pay per use; the current rates are on the pricing page.

Frequently asked questions

Is Claude free?

Yes. Free costs $0 and includes Sonnet and Haiku with lower usage limits than the paid plans.

Is Claude Pro worth it?

If you use Opus or Claude Code on most days, yes. $20 a month covers both, or $17 a month on annual billing. If you send a handful of requests a week, the API costs less.

Does Claude Pro include Fable?

Not inside the plan. Pro includes Opus, Sonnet and Haiku, and Fable runs on paid usage credits at API rates. Max, Team Premium and Enterprise include Fable within the plan's limits.

Claude Pro or Max?

Pro if you rarely see a limit. Max 5x at $100 if Pro limits interrupt you every week, and Max 20x at $200 if you also want the larger API credit bundle.

Are these prices shown with tax?

No. All prices are in US dollars and exclude tax.

Sources

• Claude pricing
• Claude API pricing

Both pages checked October 24, 2026.