---
title: "Anthropic launches Claude Fable 5.2 with 20% Lower API Prices"  
description: "Anthropic's Claude Fable 5.2 cuts API prices to $8/$40 per million tokens, keeps a 1M context window and adds fast mode. Specs and benchmarks."  
author: "Yogendra  Mohan"  
published: 2026-11-12  
updated: 2026-11-12  
canonical: https://www.mindstick.com/news/8009/anthropic-launches-claude-fable-5-2-with-20-lower-api-prices  
category: "artificial intelligence"  
tags: ["artificial intelligence", "Anthropic", "LLMs"]  
reading_time: 3 minutes  

---

# Anthropic launches Claude Fable 5.2 with 20% Lower API Prices

Anthropic released **Claude Fable 5.2** on November 9, 2026. It is the company's top model for long, demanding reasoning and agent work. It arrived alongside **Claude Mythos 5.2**, a variant restricted to vetted organizations.

## What it is for

Anthropic describes Fable 5.2 as built for agent sessions that run for hours, multistep research, and analysis that ends in a finished document, spreadsheet or deck. The API model ID is `claude-fable-5-2`.

## Specifications

- **Context window:** 1 million tokens, with up to 128,000 tokens of output.
- **Input and output:** text and images in, text out.
- **Knowledge cutoff:** June 2026.
- **Thinking:** adaptive thinking is always on and can't be turned off. Developers steer it with the effort setting, which defaults to `high`.
- **Availability:** the Claude API, Amazon Bedrock, Google Cloud, Microsoft Foundry and Claude Platform on AWS. On the Claude apps it is open to Pro, Max, Ultra, Team and Enterprise users.

## Pricing

List prices fall about 20% from Fable 5.1.

|  | Fable 5.1 | Fable 5.2 |
| --- | --- | --- |
| Input per million tokens | $10 | **$8** |
| Output per million tokens | $50 | **$40** |
| Cache read per million tokens | $0.25 | **$0.20** |

The Batch API takes 50% off. Fast mode, which Anthropic says can speed up token generation by up to 2.5 times, costs double the standard rate at $16 input and $80 output. Anthropic also says 5.2 generates tokens up to 30% faster than 5.1. For comparison, Opus 5.5 costs $4 and $20.

![Anthropic launches Claude Fable 5.2 with 20% Lower API Prices](https://www.mindstick.com/usernews/674774/62c2e012-3eb3-4c21-88b8-f2e8c83b4894/images/767e64c9-ea0a-4680-a144-9871013954e1.jpg)

## Benchmarks (Anthropic's own results)

Per the launch comparison, as reported by VentureBeat:

- **FrontierCode 2.0:** Fable 5.2 scored 49.1%, against 45.5% for Opus 5.5 and 42.8% for GPT-6 Astra.
- **FrontierSWE v2:** 69.9%, against 65.5% for Astra.
- **OSWorld 2.1:** 54.6%, against 53.2% for Opus 5.5 and 47.6% for Fable 5.1.

The results are not all one-sided. On Zapier's AutomationBench, Google's Gemini 4 Carbon scored higher, 53.9% against 51.4%. Cost also climbs steeply with effort. On FrontierCode 2.0, a task at the Max setting averaged $44.70, but scored below the cheaper Xhigh setting ($14.19 per task).

## What developers need to know

- Forcing a specific tool with `tool_choice` set to `any` or `tool` returns a 400 error.
- Leave out `temperature`, `top_p` and `top_k`.
- The model requires 30-day data retention, so it is not available under zero data retention unless Anthropic expressly authorizes it.
- Anthropic says it will not retire the model before November 9, 2027.

## Safety and Mythos 5.2

Mythos 5.2 offers the same capabilities, with different safeguards, to organizations verified through Anthropic's Cyber and Life Sciences programs. Anthropic says both models went through safety testing that included outside evaluators. Blocked cybersecurity and biology requests can fall back to older Claude models.

![Anthropic launches Claude Fable 5.2 with 20% Lower API Prices](https://www.mindstick.com/usernews/674774/62c2e012-3eb3-4c21-88b8-f2e8c83b4894/images/ba332749-8cb7-4c4b-8f47-a05258f4484d.jpg)

---

Original Source: https://www.mindstick.com/news/8009/anthropic-launches-claude-fable-5-2-with-20-lower-api-prices

Copyright © MindStick Software Pvt. Ltd. This Markdown version is provided for developers, AI systems, and offline reading.