Claude Pro

At a daily user's volume the subscription costs less than the same conversations would through the vendor's own API, so the plan is priced below its own inference bill.

FAIR

The price is mostly buying things that are not the model: evals, data pipelines, integrations, compliance, support, real interface work.

Vendor
Anthropic
Category
Assistant
Tier
Pro
Price / month
$20
Est. API cost
$42.75
Est. markup
0.47×
Price checked
13 Aug 2026
Verification
verified
§1 · Billing

What the price actually is

$20/month billed monthly; $17/month equivalent on an annual plan billed at $200 up front. Max tiers start at $100/month and are not covered by this entry. Usage limits apply and tighten at peak times.

§2 · Assumption

The usage this is priced against

A daily user with 150 substantial conversations a month, each carrying roughly 15,000 tokens of accumulated context and returning 800 tokens, answered by a large frontier model.

Someone using it half as much sees double the multiple. The assumption is the argument — if you disagree with it, the number below is not about you. Change it in the calculator.

§3 · Arithmetic

The cost math, in full

Estimated monthly inference cost

2,250,000 input tokens$33.75

Large frontier text model at $15.00 / 1M input tokens · 150 conversations; context grows through a thread, so input dominates

120,000 output tokens$9.00

Large frontier text model at $75.00 / 1M output tokens

Estimated cost$42.75
Price paid (Pro)$20
Estimated markup0.47×

Rates come from a dated rate card of representative published API prices, not from the vendor. Nobody outside these companies knows what they actually pay; volume discounts and in-house serving both push real costs below these figures, which makes every multiple here a floor rather than a ceiling.

§4 · Reasoning

Why this verdict, not the number

There is no wrapper here to price: the subscription buys access to the model the vendor trains, from the vendor. The arithmetic is only interesting for what it says about flat pricing — a heavy user is subsidised by lighter ones, and the published API rate for the same conversations exceeds the fee. Read the multiple as a statement about the included allowance rather than about the company's economics, which the usage limits exist to manage.

What the price buys besides tokens

  • Proprietary model
  • Scale & infra
  • UX craft

The weights are trained in-house, so the model is the product rather than a layer over one. Around it sits serving capacity, long-context handling, and the interface work that turns a completion endpoint into something usable without an API key.

§5 · Leaving

What you lose if you leave

The specific model, its context handling, and the ability to use it without managing keys or a client. Nothing about this is reproducible with someone else's API, because the model is the thing you are buying.

The honest cheaper path

Use the API directly with any client. At light usage that is cheaper than the subscription; at the volume assumed here it is estimated to cost roughly twice as much, and you take on rate-limit handling and a client of your own.

§6 · Your numbers

Recompute it for yourself

Recompute this for your own usage

Your estimated cost
Your markup
Cost per unit of our assumption

Scaling assumes your usage has the same shape as ours, just more or less of it. If your mix is different — far more output than input, say — the estimate drifts. It is an estimate either way.

§7 · Sources

Where every number came from

Price recorded 13 Aug 2026 · entry last reviewed 13 Aug 2026. Think something here is wrong? File a correction — we publish them, including the ones that embarrass us.