Anthropic

New

Claude Fable 5 pricing

Anthropic's June 2026 flagship and the first model in the Claude 5 family. A Mythos-class model that sits above the Opus tier, built for autonomous knowledge work and coding.

Input

$10.00/ 1M tok

Output

$50.00/ 1M tok

Context window
1M
Max output
128K
Cached input
$1.0000 / 1M
Verified
2026-07-01

Prompt caching rates

TierInput / 1MOutput / 1MCached in / 1M
Cache read (hit / refresh)
90% off standard input
$1.00$50.00-
Cache write (5-minute)
1.25x standard input
$12.50$50.00-
Cache write (1-hour)
2x standard input
$20.00$50.00-

Output is always billed at the standard $50 / 1M rate; caching only changes how input tokens are priced. Cache reads cost $1.00 / 1M (a 90% saving), a 5-minute cache write costs $12.50 / 1M, and a 1-hour cache write costs $20.00 / 1M.

Claude Fable 5 is Anthropic's June 2026 flagship, released on 9 June 2026 as the first model in the new Claude 5 family. It is a Mythos-class model – a new tier that sits above Opus in capability – and is positioned as Anthropic's most intelligent generally available model, built for autonomous knowledge work and coding. Claude Fable 5 and Claude Mythos 5 share the same underlying model; Fable 5 ships with additional safety measures for dual-use capabilities, while Mythos 5 is available without those measures to approved organisations only.

Pricing is $10 per 1M input tokens and $50 per 1M output tokens – double the Opus 4.8 rate card, reflecting the step up in capability. Prompt caching is where the effective cost can drop sharply: cache reads bill at just $1.00 / 1M (a 90% saving on input), so workloads with large, stable prefixes (long system prompts, few-shot templates, big tool definitions) can run far below the headline rate. A 5-minute cache write costs $12.50 / 1M and a 1-hour write costs $20.00 / 1M, so caching pays off once a prefix is reused enough times to amortise the write.

Context is 1M tokens by default with up to 128K output per request. If you are choosing between Fable 5 and Opus 4.8, the question is whether the extra capability justifies the 2x per-token cost for your workload; for the hardest autonomous and coding tasks Fable 5 is the new top of the range.

Estimate LLM costs before you send

Paste your prompt into the Calcis estimator to see token counts and per-request cost across every tracked model, then compare Claude Fable 5 against them side by side.

Frequently asked

How much does Claude Fable 5 cost per request?
At standard rates a 1,000-token prompt with a 500-token reply runs about $0.035 ($0.01 input + $0.025 output). With a cached prefix the input side drops to a tenth of its cost, so heavily-cached workloads run well below that.
Is Claude Fable 5 more expensive than Opus 4.8?
Yes. Fable 5 is $10/$50 per 1M tokens versus Opus 4.8's $5/$25 - roughly double. Fable 5 is a higher-capability Mythos-class model positioned above the Opus tier.
What is the difference between Fable 5 and Mythos 5?
They share the same underlying model and the same $10/$50 pricing. Fable 5 is the generally available version with additional safety measures for dual-use capabilities; Mythos 5 ships without those measures and is available only to approved organisations.
How much does prompt caching save on Fable 5?
Cache reads bill at $1.00 / 1M input tokens, a 90% discount on the $10 standard rate. Cache writes cost $12.50 / 1M (5-minute) or $20.00 / 1M (1-hour). Output is unaffected at $50 / 1M.
What is the context window for Claude Fable 5?
1M tokens by default, with up to 128K output tokens per response.

Pricing verified 2026-07-01 from the provider's rate card. These figures are informational and not yet wired into the Calcis estimator or billing.