How to try Claude Fable 5.2 this week: who gets it, what it costs and how to switch it on

Anthropic released Claude Fable 5.2 and Claude Mythos 5.2 on 9 November 2026. Fable 5.2 is open to everyone, while Mythos 5.2 is restricted. This guide sets out, step by step, how to get to the model, what it costs and which settings matter. Everything below comes from Anthropic's launch page and CNBC's report, and Anthropic's exact plan details can change, so check your own account before you rely on them.

Step 1: Decide whether you need Fable 5.2 at all

Anthropic's own guidance is that most people should start with Claude Opus 5.5, which it calls "our strongest daily driver", and switch up to Fable or Mythos only when the work is harder, or down to Sonnet 5.5 or Haiku 5.5 for simpler jobs. Dianne Penn, Anthropic's head of product management for research, told CNBC that Fable 5.2 is for "power users or power use cases". If your work is everyday writing or routine code, test Opus first and keep Fable for the problems it cannot crack.

Step 2: Check whether your plan includes it

Anthropic says plans and seats that include Fable models get one to three Fable resets, depending on how long you have subscribed, and that Fable is included for up to 50% of weekly usage limits for Pro plans and standard seats on the Team plan. It calls this a launch promotion running this week. The company's launch page links to the full terms, so read those before assuming your seat qualifies.

Step 3: Developers, call the model by its ID

On the API side, Anthropic says developers can get started with the model ID claude-fable-5-2 on the Claude Platform. It is also available on Amazon Web Services, Google Cloud and Microsoft Azure. If your company already buys Claude through one of those clouds, check that cloud's model catalogue rather than creating a new account.

Step 4: Pick the right effort setting

Effort controls how hard the model works on a task. Anthropic says Medium is the default in the Claude apps and Claude Code, and that at Medium Fable 5.2 can handle challenging tasks with fewer tokens than Fable 5.1. In early FrontierCode 2.0 testing, Fable 5.2 at High effort scored about level with Opus 5.5 at Max effort for roughly half the cost per task, according to Anthropic, which also says Xhigh gave the highest score of any released model it tested. These are the company's own results, so test on your own tasks.

• Start at Medium for most work, since it is the default and uses fewer tokens than Fable 5.1.
• Move to High when answers are close but not good enough.
• Use Xhigh only for the hardest problems, where Anthropic's own benchmarks show the best scores.

Step 5: Decide whether fast mode is worth double the price

Fast mode is optional and speeds up token generation by up to 2.5 times. It costs $16 per million input tokens and $80 per million output tokens, double the standard $8 and $40. Anthropic says it is available on Claude Code, the Claude Platform, Claude Models on Google Cloud and Microsoft Foundry. Mythos fast mode is limited to organisations in the Verification Programs. Use it for interactive work where waiting costs more than tokens, and leave it off for batch jobs.

Step 6: Know what Fable will and will not answer

Fable is the version with strong cybersecurity and biology safeguards. When a query is blocked, Anthropic says users get a reply from Opus 4.8 for cybersecurity or Opus 5 for biology instead. The company reports about 55% fewer blocks than Fable 5.1 in early testing, around 60% fewer in cybersecurity and 40% fewer in biology. For dual-use work, vetted organisations can apply for Mythos 5.2 through the Cyber Verification Program or the Life Sciences Verification Program.

Step 7: Use Fable as an advisor or orchestrator

Anthropic suggests a cheaper way to use the model. Fable 5.2 can play the role of an Advisor, where Haiku, Sonnet or Opus models consult it to solve harder problems, or an Orchestrator, where it delegates simpler tasks to smaller models. In practice that means the expensive model is called only when it is needed, while routine steps run on cheaper ones. Teams that watch their token bill should try this pattern before sending every request to the flagship.

A sensible first test is to pick three real tasks from your own work, run each on Opus 5.5 and on Fable 5.2 at Medium effort, and compare the quality and the cost of each finished task, not just the price per token. That tells you quickly whether the upgrade is worth it for you.

What does Claude Fable 5.2 cost?

Standard pricing is $8 per million input tokens and $40 per million output tokens, which Anthropic describes as lower than before and CNBC puts at 20% lower than the 5.1 generation. Anthropic says other customers have 30-day retention for safety monitoring, a point worth raising with your legal or security team before moving sensitive workloads. For a wider comparison, see our explainer on Fable 5.2 vs Opus 5.5.