Anthropic Just Launched Claude Fable 5.2 and Mythos 5.2: 30% Faster, Cheaper, and Built for the Hardest Technical Work

Anthropic announced today, November 9, 2026, two new models: Claude Fable 5.2 and Claude Mythos 5.2. They succeed Fable 5.1 and Mythos 5.1 from September, and Anthropic is positioning them for what it calls "the most ambitious, open-ended technical work" — agentic coding, cybersecurity, scientific research, and long-horizon reasoning. The headline numbers: outputs up to 30 percent faster than the 5.1 versions, fewer tokens used on hard tasks, and lower prices — $8 per million input tokens and $40 per million output tokens.

I have not run these models yet — nobody outside Anthropic had access before today — and this article is built from Anthropic's announcement and the Reuters and MacRumors coverage of it, checked 9 November 2026. Where a claim is Anthropic's, I'll say so, because a launch-day benchmark is a vendor claim, not an independent result.

What actually changed

Fable 5.2 and Mythos 5.2 are the same underlying model, split into two versions by safeguards. Fable 5.2 is the public one, with cybersecurity and biology safeguards built in, and it is available today on all platforms — including AWS, Google Cloud and Microsoft Azure. Mythos 5.2 drops those limitations, which is exactly why you probably cannot use it: access is restricted to organisations approved through Anthropic's Life Sciences and Cyber Verification Programs, plus Claude Security on Enterprise plans.

For most developers, then, the launch is really about one model: Fable 5.2. And the changes Anthropic lists for it are practical rather than flashy. Outputs arrive up to 30 percent faster than Fable 5.1. On challenging tasks at medium effort, Fable 5.2 uses fewer tokens than its predecessor — which matters because fewer tokens is not just speed, it is money on every API call. Anthropic also says Fable 5.2 has roughly 55 percent fewer blocks than Fable 5.1, which in plain terms means it gets stuck less often and needs less hand-holding on hard problems. And there is a fast mode on Claude Code, the Claude Platform, Claude Models on Google Cloud and Microsoft Foundry that speeds up token generation by up to 2.5 times.

The price moved in the right direction too: $8 per million input tokens and $40 per million output tokens, lower than the pricing for the previous Fable and Mythos releases. A faster model that also costs less per token is the combination developers actually notice — speed alone is a demo, speed plus a lower bill is a reason to switch a workload.

Where Anthropic says it beats Opus 5.5

This is the part worth reading carefully, because Anthropic is not shy about it. In its own pre-release tests, Fable and Mythos 5.2 outperform Opus 5.5 — the model Anthropic itself recommends as the starting point for most coding and knowledge work — on computer use, knowledge work, cybersecurity, agentic scientific research, agentic coding, and multidisciplinary reasoning. One specific number Anthropic published: in its internal RoboDojo Sim test for a two-armed robot, Fable 5.2 scored 64.9 percent at the highest effort level.

The usage guidance is unusually honest for a launch announcement. Anthropic recommends starting with Opus 5.5 for most coding and knowledge tasks, then switching to Fable or Mythos when a request gets complex enough to need them, or stepping down to Sonnet or Haiku 5.5 when the work is cheaper. That is a routing ladder, not a "use our newest model for everything" pitch — and it tells you something about where Fable 5.2 is meant to sit: not the default, but the model you reach for when the task is genuinely hard.

What it costs you to try it

Fable 5.2 is available today, and Anthropic is giving paid-plan subscribers extra Fable usage this week — one to three Fable resets depending on subscription length. Fable is included for up to 50 percent of weekly usage limits for Pro plans and standard seats on the Team plan. If you are already paying for Claude, the cheapest way to find out whether Fable 5.2 earns its place in your workflow is to point it at the hardest task you have open right now and watch whether it finishes without the usual mid-task stall.

Mythos 5.2, by contrast, is not something to plan around unless your organisation is already in Anthropic's verification programs. The restriction is the point: an unrestricted frontier model for cybersecurity and life-sciences research is not handed out on a pricing page.

The catch, stated plainly

• Every performance claim is Anthropic's own. The 30 percent speed gain, the token savings, the wins over Opus 5.5 — all from Anthropic's pre-release tests, none independently replicated yet. Treat them as a reason to try the model, not as a settled fact.
• Fable 5.2 is for the hard tasks, not every task. Anthropic's own guidance says start with Opus 5.5 and step down to Sonnet or Haiku 5.5 for cheaper work. Using Fable for everything would be the expensive way to follow this launch.
• Mythos 5.2 is effectively unavailable to most developers. If you are not in a verification program or on an Enterprise plan with Claude Security, the unrestricted version is not on offer, and no pricing page changes that.
• The IPO context is real. Anthropic filed its registration statement on October 26 and plans to list on Nasdaq under the ticker ANT. A launch two weeks before an IPO is still a real product launch — but it is also a company showing its strongest hand at the moment it most wants to be seen holding one.

The bottom line

Fable 5.2 is not a new kind of model. It is the same idea Anthropic has been building toward all year — a model for long, open-ended technical work — made faster, cheaper per token, and less likely to stall. The 30 percent speed gain and the lower price are the parts that will show up in a real workflow first; the benchmark wins over Opus 5.5 are the parts that need independent testing before anyone should quote them.

If you already use Claude Code or the Claude API, the honest test costs you nothing extra this week: give Fable 5.2 the hardest task on your plate — the one Opus 5.5 usually needs two or three attempts to finish — and see whether it gets there in one.

Would you switch your hardest coding task to Fable 5.2 this week, or wait for independent benchmarks first?

Sources: Anthropic announcement, 9 November 2026; Reuters — "Anthropic unveils new Fable, Mythos models as competition heats up ahead of IPO" (https://www.reuters.com/technology/anthropic-unveils-new-fable-mythos-models-competition-heats-up-ahead-ipo-2026-11-09/), 9 November 2026; MacRumors — "Anthropic Launches Claude Fable 5.2 and Mythos 5.2" (https://www.macrumors.com/2026/11/09/anthropic-launches-claude-fable-5-2/), 9 November 2026. All capability, benchmark and pricing claims are Anthropic's, stated as such. No hands-on testing was performed for this article. This breaking article is not the daily buyer-intent post, and it contains no affiliate links.