How much does Claude Fable 5.2 cost on Hormouz?

Until November 16, 2026, 40% off: $4.80 per million input tokens and $24 per million output tokens. After that, Anthropic's list price of $8 and $40.

Is Fable 5.2 better than Gemini 4 Carbon?

On the Artificial Analysis Intelligence Index, yes (64.6 against 59), and it is stronger on long agent work. Carbon is cheaper and writes Persian in fewer tokens; the comparison page has every number side by side.

Does it work from Iran without an Anthropic account?

Yes. You send requests with a Hormouz key and Hormouz carries them to Anthropic; no Anthropic account, foreign card or VPN is needed.

Does it work with Claude Code or Anthropic's own SDK?

No. Hormouz has an OpenAI-compatible API: the OpenAI SDK and compatible tools work with the model name anthropic/claude-fable-5.2, but Claude Code and Anthropic's own SDK do not connect.

Fable 5.2 or Opus 5.5?

Fable 5.2 is stronger on the tests; Opus 5.5 is cheaper and enough for lighter work. One key reaches both, so you can send the hard work to Fable.

Claude Fable 5.2 in Iran

The newest and strongest Claude model, on Hormouz from its release week. One Hormouz key and rial payments, with no Anthropic account and no monthly subscription.

Create an account Full model details

Pay in rial · No monthly subscription · Usage at Google's official rate

What is Claude Fable 5.2 for?

Anthropic released Claude Fable 5.2 on November 9, 2026. It is built for heavy reasoning and long agent work: sessions that run for hours, multi-step research, and analysis carried through to a finished document, spreadsheet or deck. Anthropic reports it beats Fable 5.1 at coding, reasoning and vision, and writes about 30% faster.

It reads text and images, has a 1-million-token context window, and one reply can run to 128,000 tokens. It always thinks before answering, and the reasoning effort setting (low to max) sets how much.

Fable 5.2 at the top

Artificial Analysis Intelligence Index; higher means the model did better on reasoning, coding and knowledge tests.

Claude Fable 5.2 price on Hormouz

	Input (per million tokens)	Output (per million tokens)
Until November 16, 2026	$4.8/M	$24/M
After November 16, 2026	$8/M	$40/M

The Hormouz discount applies to requests served over our direct Anthropic connection, the normal path for this model; a request that goes over the backup path is billed at list price, and the invoice shows which. Reasoning tokens count as output.

Automatic caching cuts the cost of repeated prompts

Hormouz sets a cache breakpoint on every Fable 5.2 request; when the start of a prompt (instructions or a long document) repeats, it is billed at the cache-read rate from the second call on, with no change to your code.

Example: Claude Fable 5.2 with a fixed 50,000-token prompt, called 20 times, 1,000 output tokens each.

Without caching: $8.80
With caching: $1.49
Saving: 83%

Input: $8/M
Cache write: $10/M
Cached input: $0.2/M

Claude Fable 5.2 against its rivals

Price, context window, capabilities and the cost of Persian text, side by side.

Give your hardest work to the strongest Claude.

A new account comes with gift credit, and after that you pay only for what you use.

Create an account

FAQ

How much does Claude Fable 5.2 cost on Hormouz?: Until November 16, 2026, 40% off: $4.80 per million input tokens and $24 per million output tokens. After that, Anthropic's list price of $8 and $40.
Is Fable 5.2 better than Gemini 4 Carbon?: On the Artificial Analysis Intelligence Index, yes (64.6 against 59), and it is stronger on long agent work. Carbon is cheaper and writes Persian in fewer tokens; the comparison page has every number side by side.
Does it work from Iran without an Anthropic account?: Yes. You send requests with a Hormouz key and Hormouz carries them to Anthropic; no Anthropic account, foreign card or VPN is needed.
Does it work with Claude Code or Anthropic's own SDK?: No. Hormouz has an OpenAI-compatible API: the OpenAI SDK and compatible tools work with the model name anthropic/claude-fable-5.2, but Claude Code and Anthropic's own SDK do not connect.
Fable 5.2 or Opus 5.5?: Fable 5.2 is stronger on the tests; Opus 5.5 is cheaper and enough for lighter work. One key reaches both, so you can send the hard work to Fable.

Prices updated: 2026-11-11