Claude Fable 5.2 is Anthropic’s newest top-tier AI model, and it arrived on November 9, 2026 alongside a restricted sibling, Claude Mythos 5.2. The update promises output up to 30% faster, cheaper API pricing and far fewer cases where the model refuses a harmless request by mistake.
Claude Fable 5.2 is the third Fable release in five months, after Fable 5 in June and Fable 5.1 on September 1, and it lands just as Anthropic prepares for a widely expected stock market debut.
In this article
What is Claude Fable 5.2?
Fable and Mythos share the same underlying model. The difference is the guardrails. Claude Fable 5.2 is the public version, with extra safeguards around cybersecurity and biology that limit how it answers high-risk requests. Claude Mythos 5.2 lifts some of those limits and is only offered to vetted organizations.
Anthropic says both models improve on coding, reasoning and vision, and that they explain their work more clearly than the 5.1 versions. In one evaluation published with the launch, Fable 5.2 at Max effort scored 83.5, compared with 65.3 for Opus 5.5 and 44.4 for Fable 5.1. That is a big jump for a point release, though it is the company’s own test.
Claude Fable 5.2: the 5 biggest changes
| Change | What Anthropic says |
|---|---|
| Speed | Outputs generated up to 30% faster than the 5.1 models |
| Fast mode | Up to 2.5x faster token generation on supported platforms |
| Efficiency | Fewer tokens than Fable 5.1 on hard tasks at medium effort |
| Fewer false blocks | About 55% fewer blocks than Fable 5.1, with 60% fewer in cybersecurity and 40% fewer in biology |
| Price | $8 per million input tokens and $40 per million output tokens, down from $10 and $50 |
The drop in blocks may matter most to everyday users. A safety classifier that misfires stops a legitimate question cold, and Anthropic says it has kept working on those false positives since 5.1. It measured the improvement in early testing across its Claude apps and Claude Code.
A customer example on Anthropic’s Fable page shows the kind of work Claude Fable 5.2 targets. A materials science team says Fable 5.2 made the phonon code in Quantum ESPRESSO about 36 times faster on the same GPU, cutting 13 benchmark cases from 3.3 hours to under four minutes. Treat that company-picked showcase as a best case, not a typical result.
Fable 5.2 pricing and availability
Claude Fable 5.2 is available now on all of Anthropic’s platforms, including the Claude apps and the Claude Platform API, plus Amazon Web Services, Google Cloud and Microsoft Azure. Developers can call it with the model ID claude-fable-5-2.
The new price is 20% below Fable 5.1 on both input and output tokens. Fortune reported that the pricing also undercuts OpenAI’s GPT-6 Astra, which came out in early September.
Paid Claude subscribers on eligible plans get extra Fable usage this week, with one to three usage resets depending on how long they have subscribed, according to MacRumors. Fast mode is offered on Claude Code, the Claude Platform, Claude models on Google Cloud and Microsoft Foundry.
Who can use Claude Mythos 5.2?
Mythos 5.2 is not a general release. Anthropic offers it through its Life Sciences and Cyber Verification Programs, and Enterprise customers can reach it through Claude Security. The company has kept Mythos-class access tight since the first preview in April, arguing that its cybersecurity skills need closer control.
Anthropic also flagged distillation, where other developers train their own models on Claude’s output. It says this can strip away the safeguards it built and creates safety and national security risks, which is part of why Fable keeps its guardrails. Lawmakers are debating limits on frontier AI too, as we covered in our look at the Ban Artificial Superintelligence Act.
Why the Claude Fable 5.2 release matters
Timing is a big part of the story. CNBC reports the launch comes as Anthropic gets ready for an IPO that could happen as soon as next week. A cheaper, faster flagship gives the company a fresh story for investors and a sharper pitch against OpenAI and Google.
For regular users, the practical gains are speed and fewer false refusals. For developers, a 20% price cut adds up quickly on long agent runs that burn millions of tokens. CNBC also notes that Anthropic still points people to Opus 5.5 for everyday tasks and to Fable for hard problems in engineering, math and robotics.
Independent tests over the next few weeks will show whether the Claude Fable 5.2 gains hold up outside Anthropic’s own evaluations. Follow more AI news on our latest stories page, or read how ChatGPT Images 2.6 added GIF support.
FAQ
Is Claude Fable 5.2 free?
Anthropic’s announcement focuses on eligible paid Claude plans and the paid API. Check the model picker in your Claude app to see whether your plan includes it.
How much does Claude Fable 5.2 cost on the API?
It costs $8 per million input tokens and $40 per million output tokens, down from $10 and $50 for Fable 5.1.
What is the difference between Fable 5.2 and Mythos 5.2?
They share one model. Fable 5.2 is public and has extra cybersecurity and biology safeguards, while Mythos 5.2 is limited to vetted programs and Enterprise customers.
Sources: Anthropic, Anthropic Fable page, CNBC, MacRumors, Fortune, Yahoo Finance
Leave a Reply