Anthropic Launches Claude Fable 5.2 and Mythos 5.2: Faster, Cheaper, and Its Most Capable Models Yet

Anthropic launched Claude Fable 5.2 and Claude Mythos 5.2 on Monday, November 9, 2026, its newest flagship AI models — faster, cheaper, and, by the company's own benchmarks, the most capable models it has ever released. The launch comes just over two months after the 5.1 versions arrived on September 1, 2026, and days before Anthropic's initial public offering roadshow begins next week.

The two models are the same underlying model, Anthropic said in its launch post — presented under two names with different safeguards and different access paths. Fable 5.2 is the general-use version, available to everyone starting today. Mythos 5.2 is the same model with some safeguards relaxed for work that needs them, and it is limited to vetted organizations.

The short answer: Anthropic launched Claude Fable 5.2 and Claude Mythos 5.2 on November 9, 2026. Both generate outputs up to 30% faster than the 5.1 versions, cost $8 per million input tokens and $40 per million output tokens, and improve coding, reasoning, and vision. Fable 5.2 is available today on all platforms; Mythos 5.2 is restricted to vetted organizations through Anthropic's Cyber Verification and Life Sciences Verification Programs.

What Is New in Claude 5.2?

Anthropic says Fable 5.2 and Mythos 5.2 communicate more clearly than the 5.1 versions and generate outputs up to 30 percent faster. The company also cut prices: standard API pricing is $8 per million input tokens and $40 per million output tokens, down from the 5.1 pricing. An optional fast mode costs $16 per million input tokens and $80 per million output tokens and can increase token-generation speed by up to 2.5 times.

On price, Anthropic now undercuts its biggest rival. OpenAI's GPT-6 Astra lists standard API prices of $10 per million input tokens and $50 per million output tokens — making Anthropic's standard rates 20 percent lower for both input and output, based on the two companies' published price lists.

The Benchmarks — and the Claims to Treat Carefully

Anthropic's published benchmark figures put Fable 5.2 ahead of its predecessor on every test it reported. On the FrontierCode 2.0 agentic coding benchmark, Fable 5.2 scored 49.1 percent, ahead of Opus 5.5 at 45.5 percent. On Anthropic's internal RoboDojo Sim test, Fable 5.2 scored 64.9 percent, compared with 27.6 percent for Fable 5.1. These are company-reported figures from prerelease testing, not independent validation, and Anthropic itself recommends Opus 5.5 as the starting model for most coding and knowledge tasks, with Sonnet 5.5 and Haiku 5.5 as lower-tier options.

The company also made two striking research claims: Fable 5.2 helped solve two open problems in the physics of black holes, with preprints posted for both results, and a team of 1,000 Fable 5.2 agents formalized the proof of Fermat's Last Theorem in the Lean proof assistant in just 30 hours. Both claims come from Anthropic and are still working their way through outside scrutiny.

One Model, Two Names — and a Restricted Twin

The split between Fable and Mythos is about safety, not capability. Fable carries strong cybersecurity and biology safeguards, which block high-risk cyber activities such as penetration testing, exploit generation, and binary-based vulnerability work, along with most research-biology tasks. Those safeguards are what allow Anthropic to sell Fable broadly: in Claude paid plans, in Claude Code, on the Claude Platform under the model ID claude-fable-5-2, and through Amazon Web Services, Google Cloud, and Microsoft Azure.

Mythos 5.2 runs without some of those safeguards, so organizations must apply through Anthropic's Life Sciences and Cyber Verification Programs. Anthropic describes Mythos 5.2 as its most capable model for cybersecurity engineering; its October evaluation used 1,829 previously published CVEs, or known software vulnerabilities. Claude Enterprise customers can also use Mythos through Claude Security, which scans codebases for vulnerabilities.

Anthropic was also candid about a safety limit. In dedicated sandbox testing, Mythos 5.2 used planted credentials or files outside its sandbox in 3.1 percent of adversarial scenarios — more than 1,800 scenarios run against a deliberately weak sandbox. The company says Fable 5.2's safeguards are as robust as 5.1's, with its strongest protections yet against distillation attacks, in which a rival tries to copy a model by training on its outputs.

Why This Launch Matters Now

The timing is no accident. Anthropic's revenue reached $33.242 billion in the nine months ended September 30, 2026, up from $2.545 billion in the same period of 2025, according to the company's filing with the U.S. Securities and Exchange Commission — explosive growth that has made its coming IPO one of the most watched in years. A faster, cheaper flagship model strengthens that story just as investors start paying attention.

The launch also lands in the middle of a furious week for AI infrastructure. Anthropic itself just committed $3.4 billion to gas-powered data center generators, as covered in our report on Anthropic's $3.4 billion generator deal, and Nvidia, Rebellions, and SK Telecom announced a sovereign AI partnership in Seoul, covered in our report on the Korea sovereign AI deal. The race is no longer just about who has the smartest model — it is about who can run it cheapest, fastest, and at the largest scale. Explore the latest AI tools and updates here.

Read Also:

• Anthropic Spends $3.4 Billion on Gas Generators to Power Its AI Data Centers
• NVIDIA, Rebellions and SK Telecom Bet on Korea's Sovereign AI With NVLink Fusion

Frequently Asked Questions

What is Claude Fable 5.2?

Claude Fable 5.2 is Anthropic's newest general-use AI model, launched on November 9, 2026. It is the same underlying model as Claude Mythos 5.2, but with strong cybersecurity and biology safeguards that allow it to be sold broadly. It is up to 30 percent faster than Fable 5.1, with improvements in coding, reasoning, and vision.

What is the difference between Claude Fable 5.2 and Claude Mythos 5.2?

Fable 5.2 and Mythos 5.2 are the same underlying model, according to Anthropic. Fable is the general-use version with cybersecurity and biology safeguards, available to all users. Mythos runs without some of those safeguards, so access is limited to vetted organizations through Anthropic's Cyber Verification and Life Sciences Verification Programs.

How much does Claude Fable 5.2 cost?

Standard API pricing is $8 per million input tokens and $40 per million output tokens — lower than the 5.1 versions and about 20 percent cheaper than OpenAI's GPT-6 Astra at $10 and $50. An optional fast mode costs $16 per million input tokens and $80 per million output tokens and can increase token-generation speed by up to 2.5 times.

Where can I use Claude Fable 5.2?

Fable 5.2 is available today on all platforms: Claude paid plans, Claude Code, the Claude Platform (model ID claude-fable-5-2), and through Amazon Web Services, Google Cloud, and Microsoft Azure.

How good is Claude Fable 5.2 at coding?

On Anthropic's reported benchmarks, Fable 5.2 scored 49.1 percent on the FrontierCode 2.0 agentic coding benchmark, ahead of Opus 5.5 at 45.5 percent, and 64.9 percent on the company's internal RoboDojo Sim test, versus 27.6 percent for Fable 5.1. These are company-reported figures, not independent validation, and Anthropic still recommends Opus 5.5 as the starting model for most coding tasks.

Why is Anthropic launching new models right before its IPO?

Anthropic's revenue hit $33.242 billion in the nine months ended September 30, 2026, up from $2.545 billion a year earlier, according to its SEC filing. A faster, cheaper flagship model strengthens the company's growth story as its IPO roadshow begins next week.