How much does Claude Fable 5.2 cost on Hormouz?

$4.80 per million input tokens and $24.00 per million output tokens — Anthropic's list price of $8.00 and $40.00 less Hormouz's 40% discount. A cache read is $0.12 per million tokens and a 5-minute cache write $6.00. Billing is per token with no monthly fee, and you are charged only for what a request actually used.

Is the 40% discount applied to every request?

It applies to every request Hormouz serves over its own direct Claude connection, which is the normal path for this model. If a request cannot go that way — an option that path does not accept, or an outage — Hormouz still answers it, but that request is billed at the list price of $8.00 and $40.00. Your invoice shows the list price and the discount as separate lines, so you can always see which price a request paid.

How do I access Claude Fable 5.2 from inside Iran?

Sign up at Hormouz, top up in rial, and call the model with a Hormouz API key. No foreign credit card, no Anthropic account and no circumvention tooling are needed — Hormouz handles the connection and serves the model from an Iran-reachable endpoint.

Is Claude Fable 5.2 compatible with OpenAI code?

Yes. Hormouz exposes it on an OpenAI-compatible chat completions API, so any OpenAI SDK or tool that lets you set a base URL and an API key works unchanged — set the model to anthropic/claude-fable-5.2. Note that the Anthropic SDK and Claude Code speak Anthropic's own protocol and do not connect to Hormouz.

What context window and output limit does Claude Fable 5.2 have?

The context window is 1M tokens, about 555,000 words, and a single reply can be up to 128,000 tokens — the model's own limit. Reasoning tokens come out of the same budget, because the model always thinks before it answers.

What is Claude Fable 5.2 good for?

Work that is long and hard rather than short and quick: agent sessions that run for hours, long code refactors, multistep research, financial and legal analysis, and reports or spreadsheets carried through to a finished artefact. For short, high-volume or latency-sensitive calls a cheaper Claude model is usually the better buy.

What changed from Claude Fable 5.1?

It is cheaper and faster: the list price dropped from $10/$50 to $8/$40 per million tokens, and Anthropic reports output generated about 30% faster, with gains in coding, reasoning and vision. Context, output limit and the API shape are unchanged, so moving over is a change of model id. Claude Fable 5.1 remains available on Hormouz.

Which parameters does Claude Fable 5.2 not accept?

temperature, top_p and top_k are refused by the model, so Hormouz strips them rather than letting your request fail; use the reasoning effort setting to control depth instead. Forced tool choice (requiring the model to call a tool) is also unsupported — offer the tools and let the model decide. Thinking cannot be turned off on this model.

Claude Fable 5.2 or Claude Opus 5.5: which should I use?

Opus 5.5 is half the price ($4/$20 at list) with the same 1M context, and it is the better default for most agentic coding and knowledge work. Pick Fable 5.2 for the hardest reasoning and the longest-running agent sessions, or when your own evaluation of Opus 5.5 at a higher effort still falls short.

Claude Fable 5.2: AI model

Claude Fable 5.2 is Anthropic's newest Claude model, built for demanding reasoning and long-horizon agentic work, with a 1M-token context window. Reachable from inside Iran through Hormouz, at 40% off.

Context window:: 1,000,000 tokens
Input:: $4.8/1M tokens
Output:: $24/1M tokens
Discount:: −40%
Type:: text, image

Vision
Tools
Streaming
Long context

Thinking levels: Always thinks · max · xhigh · high · medium · low

Claude Fable 5.2 was released on 9 November 2026 and is Anthropic's most capable model. It is built for demanding reasoning and long-horizon agentic work: agent sessions that run for hours, multistep deep research, and analysis carried through to a finished document, spreadsheet or deck. Anthropic reports gains over Claude Fable 5.1 in coding, reasoning and vision, with output generated about 30% faster. It currently ranks first on Artificial Analysis's Intelligence Index, at 65.

It takes text and images as input and returns text, and supports tool calling, structured JSON output and streaming. Thinking is adaptive and always on — it cannot be switched off — so you steer how hard it thinks with the reasoning effort setting, which defaults to high. The context window is 1M tokens, roughly 555,000 words, and a single reply can run to 128,000 tokens. Its knowledge is reliable through June 2026.

Pricing is usage-based with no monthly fee. Anthropic's list price is $8 per million input tokens and $40 per million output tokens; Hormouz applies 40% off, so you pay $4.80 per million input tokens and $24.00 per million output tokens. Prompt caching is cheaper again: a cache read costs $0.20 per million tokens at list, $0.12 after the discount, and Hormouz sets a cache breakpoint on every request for you, so a repeated prompt prefix is billed at the read rate without any change to your code.

Hormouz serves Claude Fable 5.2 behind an OpenAI-compatible API: point your base URL at Hormouz, drop in a Hormouz key, and existing OpenAI SDK code runs unchanged — no foreign credit card, no circumvention tooling, and top-ups in rial. The same key reaches every other model in the catalogue, so you can fall back to a cheaper model for simple turns and keep Fable 5.2 for the hard ones.

Claude Fable 5.2 is Anthropic's most capable model for the most demanding technical work, with gains over Fable 5.1 in coding, reasoning, and vision. In prerelease testing it completed otherwise-infeasible...

Quickstart

curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-fable-5.2",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Specs

Vendor	Anthropic
Type	Text, Image
Context window	1,000,000 tokens
Pricing	Previous price: $8 / $40 per MDiscounted price: $4.8 / $24 per M−40%
API	Chat Completions
Capabilities	Streaming · Vision · Tools

Code example

cURL

curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "anthropic/claude-fable-5.2",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Python

from openai import OpenAI
client = OpenAI(api_key="ng-YOUR_API_KEY", base_url="https://api.hormouz.net/v1")
response = client.chat.completions.create(
    model="anthropic/claude-fable-5.2",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

JavaScript

import OpenAI from "openai";
const client = new OpenAI({ apiKey: "ng-YOUR_API_KEY", baseURL: "https://api.hormouz.net/v1" });
const response = await client.chat.completions.create({
  model: "anthropic/claude-fable-5.2",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

Buy the Claude Fable 5.2 API, paid in rial

Buy credit with an Iranian bank card; no foreign card or account needed.
One Hormouz key reaches Claude Fable 5.2 and nearly 700 other models, and you pay only for what you use.
No VPN and no risk of a banned account; just change the base URL to api.hormouz.net/v1.

Create an account Buy Claude API access, paid in rial

Use via Hormouz

Connect to this model and nearly 700 others with one API key, then see usage down to each request.

Request company access

FAQ

How much does Claude Fable 5.2 cost on Hormouz?: $4.80 per million input tokens and $24.00 per million output tokens — Anthropic's list price of $8.00 and $40.00 less Hormouz's 40% discount. A cache read is $0.12 per million tokens and a 5-minute cache write $6.00. Billing is per token with no monthly fee, and you are charged only for what a request actually used.
Is the 40% discount applied to every request?: It applies to every request Hormouz serves over its own direct Claude connection, which is the normal path for this model. If a request cannot go that way — an option that path does not accept, or an outage — Hormouz still answers it, but that request is billed at the list price of $8.00 and $40.00. Your invoice shows the list price and the discount as separate lines, so you can always see which price a request paid.
How do I access Claude Fable 5.2 from inside Iran?: Sign up at Hormouz, top up in rial, and call the model with a Hormouz API key. No foreign credit card, no Anthropic account and no circumvention tooling are needed — Hormouz handles the connection and serves the model from an Iran-reachable endpoint.
Is Claude Fable 5.2 compatible with OpenAI code?: Yes. Hormouz exposes it on an OpenAI-compatible chat completions API, so any OpenAI SDK or tool that lets you set a base URL and an API key works unchanged — set the model to anthropic/claude-fable-5.2. Note that the Anthropic SDK and Claude Code speak Anthropic's own protocol and do not connect to Hormouz.
What context window and output limit does Claude Fable 5.2 have?: The context window is 1M tokens, about 555,000 words, and a single reply can be up to 128,000 tokens — the model's own limit. Reasoning tokens come out of the same budget, because the model always thinks before it answers.
What is Claude Fable 5.2 good for?: Work that is long and hard rather than short and quick: agent sessions that run for hours, long code refactors, multistep research, financial and legal analysis, and reports or spreadsheets carried through to a finished artefact. For short, high-volume or latency-sensitive calls a cheaper Claude model is usually the better buy.
What changed from Claude Fable 5.1?: It is cheaper and faster: the list price dropped from $10/$50 to $8/$40 per million tokens, and Anthropic reports output generated about 30% faster, with gains in coding, reasoning and vision. Context, output limit and the API shape are unchanged, so moving over is a change of model id. Claude Fable 5.1 remains available on Hormouz.
Which parameters does Claude Fable 5.2 not accept?: temperature, top_p and top_k are refused by the model, so Hormouz strips them rather than letting your request fail; use the reasoning effort setting to control depth instead. Forced tool choice (requiring the model to call a tool) is also unsupported — offer the tools and let the model decide. Thinking cannot be turned off on this model.
Claude Fable 5.2 or Claude Opus 5.5: which should I use?: Opus 5.5 is half the price ($4/$20 at list) with the same 1M context, and it is the better default for most agentic coding and knowledge work. Pick Fable 5.2 for the hardest reasoning and the longest-running agent sessions, or when your own evaluation of Opus 5.5 at a higher effort still falls short.

Updated 2026-11-11