Guides / claude fable 5 pricing

Claude Fable 5 pricing and the cost of a cache write

Claude Fable 5 pricing is $10.00 per million input tokens and $50.00 per million output tokens, and Claude Mythos 5 bills identically at the same rate point. Cache writes cost $12.50, above the input rate. Culpa, a local-first LLM cost, margin, and forecast ledger, meters cached, uncached and cache-write tokens separately per call.

Why this happens

Fable sits at the top of the Anthropic line, ten times Haiku on input, with the same five-to-one output ratio the whole line shares. At this rate the cache-write charge stops being a footnote. A write bills $12.50 per million against a $10.00 input rate, so a cache rebuilt and never read costs 25% more than not caching. The absolute numbers are large enough that the write-to-read ratio deserves an owner rather than a default.

What this usually looks like

A handful of Fable calls outspend thousands of calls on your production model.
Spend rose after enabling caching on this tier, the signature of writes outnumbering reads.
Long-form generation runs here because it was the best model when the feature was written.
One Fable conversation sits in your top spenders despite a tiny call count.

Free, no card, no account

Run the free Cost Leak Scan

It shows your most expensive conversation before you install anything.

Run the free Cost Leak Scan Start 14-day trial

Mistakes that cost the most

Mistake	Why it hurts	Do instead
Running a whole pipeline on Fable because the final pass needs it.	Formatting and routing steps bill at $10 and $50 per million to do work a small model does identically.	Split the pipeline by step and keep Fable for the one whose output a customer actually reads.
Letting Fable produce unbounded long-form output.	At $50 per million, a thousand extra output tokens is five cents on a single call.	Cap max tokens and treat hitting the ceiling as a signal to reshape the task.
Enabling caching without watching the write-to-read ratio.	A write costs $12.50 per million here, so a cache nobody reads is 25% worse than no cache at all.	Count writes against reads. A five-minute cache clears break-even on the first read.

Run this check tonight

Count your Fable calls this month and price them individually. The list is short and usually startling.
Measure how many tokens of each prompt repeat across calls, then price that block at $10 per million.
Count cache writes against reads on this tier specifically.
Check whether any retry or fallback path can escalate into Fable.
Confirm each Fable call is doing work that genuinely needs the top tier.

A long-form step on the top tier, priced

Illustrative example

A drafting step running 5,000 times a month. Each call sends 8,000 input tokens, of which 6,000 are an identical system block, and returns 2,000 tokens.

Input: 5,000 x 8,000 = 40M tokens x $10.00 per million = $400

Output: 5,000 x 2,000 = 10M tokens x $50.00 per million = $500

Month total = $900 across just 5,000 calls, so $0.18 per call

The repeated 6,000-token block: 30M x $10.00 per million = $300 of that input

Cached at $1.00 per million on an 80% hit rate: (24 x $1.00) + (6 x $10.00) = $24 + $60 = $84, saving $216

$900 for 5,000 calls, and $216 of it disappears by caching a system block that never changes. Output is 56% of the bill on a quarter of the tokens.

Every number, with its confidence and source

Figure	What it means	Confidence	Source
$10.00 / $50.00 per million	Claude Fable 5 and Claude Mythos 5 input and output rates, identical on both	calculated	Culpa price book rows for anthropic/claude-fable-5 and anthropic/claude-mythos-5, both $0.010 and $0.050 per 1k, verified against Anthropic pricing on 2026-08-01.
$12.50 per million	Fable 5 five-minute cache write rate, above the input rate	calculated	$0.010 per 1k input x 1.25, the published Anthropic cache-write multiplier, shown on the pricing page as $12.50 per MTok.

What a generic answer can’t know

A rate card can tell you Fable is the top tier. It can't tell you what share of your input is a repeated block paying full price, or which single conversation carried a fifth of the month. That needs request-level data. Culpa keeps it on your infrastructure, where your prompts and responses stay, and counts the calls to run your plan.

Questions founders ask next

How much does Claude Fable 5 cost?

$10.00 per million input tokens and $50.00 per million output, with cache reads at $1.00. Claude Mythos 5, listed as limited availability, bills at exactly the same rate point.

What does a Fable 5 cache write cost?

$12.50 per million for the five-minute window and $20.00 for the one-hour window, at the published 1.25x and 2x input multipliers. A five-minute cache clears break-even on the first read, so only a write nobody reads loses money.

Is Fable 5 worth it over Opus?

Fable is double Opus on both sides, $10 against $5 input and $50 against $25 output. The ratio is identical at five to one, so the decision is purely whether the output quality justifies doubling that step, measured on your own task.

On your infrastructure

Culpa runs on your infrastructure. Your prompts and responses never leave it. Culpa counts calls to run your plan, and it fails open, so if it ever breaks your app keeps running.

How Culpa works

Find the culprit. Not just the total.

Your dashboard shows what you spent. It stops short of who spent it. Culpa shows the conversation, the user and the feature behind it.

Your prompts stay local.

Culpa runs on your own infrastructure. What you send to a model reaches us at no point.

Every dollar has a name.

Follow any charge to the conversation, the user, the feature and the customer behind it.

See the bill before it lands.

Cost your next feature before you ship it. You get the likely bill and the worst case, at best, median, p90 and p99.

Three steps to your first answer.

Change one base URL.

Or drop in the Python or TypeScript library.

Find your most expensive conversation.

In the first session, not the first week.

Cost your next feature before you ship it.

Base URLhttp://localhost:4545/v1Your traffic keeps flowing if Culpa ever stops.

Why the bill went up

Example dashboard

Calls traced

418,209

across 3 projects

Spend this week

$378.41

+ $182 vs last week

Failed calls

312

74% retried, and you paid for all of them

+ $182 this week traced to one culprit

Spend over 14 days

user_384report_generatorconv_91fprompt_v1894,220 tokens3 retries$6.81

Most expensive users

user_384$38.42

user_119$21.07

user_562$14.90

user_204$8.30

user_871$5.10

Next week forecast

Best$180

Median$240

p90$310

p99$395

Graded against reality. Accuracy shown as results land.

Run the free Cost Leak Scan Start 14-day trial

Keep reading

Sources: Anthropic pricing. Last reviewed 2026-08-01, rates effective 2026-07-02. Plain text version.