Breaking
USAI hardware startup Nominal triples valuation to $3B in 8 monthsDEHolstein Kiel fires coach WalterFRNvidia affair: why Jensen Huang’s affair with a Chinese spy could shake up the entire AI sectorFR“It can only benefit the country”: Emmanuel Macron congratulates Teddy Riner on his candidacy for the presidential electionUSLions-Jets trade grades: Detroit bolsters secondary with Andre Cisco, Jets acquire a late-round pickINTLHardline Trump aide Stephen Miller takes over as White House chief of staffBRSix people are arrested for drug trafficking in Barra MansaAUUniversity backflips on decision to cut 87 per cent of arts subjectsTRClaude Fable 5.2 and Mythos 5.2 have arrivedESLeo XIV brings together two and a half million people in Buenos AiresUSAI hardware startup Nominal triples valuation to $3B in 8 monthsDEHolstein Kiel fires coach WalterFRNvidia affair: why Jensen Huang’s affair with a Chinese spy could shake up the entire AI sectorFR“It can only benefit the country”: Emmanuel Macron congratulates Teddy Riner on his candidacy for the presidential electionUSLions-Jets trade grades: Detroit bolsters secondary with Andre Cisco, Jets acquire a late-round pickINTLHardline Trump aide Stephen Miller takes over as White House chief of staffBRSix people are arrested for drug trafficking in Barra MansaAUUniversity backflips on decision to cut 87 per cent of arts subjectsTRClaude Fable 5.2 and Mythos 5.2 have arrivedESLeo XIV brings together two and a half million people in Buenos Aires
BackAnthropic releases first frontier models since Amodei's call for slowdown as IPO buzz builds
Anthropic releases first frontier models since Amodei's call for slowdown as IPO buzz builds
BREAKING
CNBC22 minutes agoTech2 min read

Anthropic releases first frontier models since Amodei's call for slowdown as IPO buzz builds

Anthropic says its newest AI models generate outputs 30% faster and cost 20% less than the previous generation, as the company prepares for an IPO as soon as next week.

Quick Look

  • Anthropic released Claude Fable 5.2 and Claude Mythos 5.2, its most advanced AI models, generating outputs 30% faster and costing 20% less than the 5.1 generation.
  • The launch comes as Anthropic prepares for a potential IPO as soon as next week, targeting a $1 trillion to $1.5 trillion valuation.

AI-generated summary

Why It Matters

Anthropic first introduced Mythos in April with a limited rollout due to advanced cybersecurity capabilities, and introduced Fable two months later as a publicly available version with additional safeguards. CEO Dario Amodei called to slow down the pace of AI development in September. The company filed an extensive risk factors section in its prospectus late last month and generated $33.24 billion in revenue for the nine months ended Sept. 30, up roughly 1,200% from the same period last year.

Font size

Anthropic on Monday released two new artificial intelligence models, Claude Fable 5.2 and Claude Mythos 5.2, as the company readies for a potentially historic IPO that could take place as soon as next week.

Fable 5.2 and Mythos 5.2 are Anthropic's most advanced offerings, and the company said they generate outputs 30% faster than their predecessors. The models also cost 20% less than the 5.1 generation, and will be available for $8 per million input tokens and $40 per million output tokens, the company said.

The launch marks the first time that Anthropic has released so-called "frontier" models since its CEO, Dario Amodei, called to slow down the pace of AI development in September. Anthropic said the models went through rigorous safety and alignment testing, and that the company developed new evaluations and post-launch monitoring tools ahead of the release.

"We called for pacing, not pausing," Dianne Penn, Anthropic's head of product management for research, told CNBC in an interview.

Anthropic has to balance its safety efforts with intense commercial and competitive pressures, a tension that the company highlighted in the extensive risk factors section of its prospectus late last month. And as executives prepare to kick off Anthropic's IPO roadshow, they'll have to convince investors that their approach is the right one.

The company is aiming for a valuation of $1 trillion to $1.5 trillion in its offering, as CNBC previously reported, down from a target of $2 trillion that had been floated in recent months.

Anthropic first introduced Mythos in April, and it limited the rollout to a select group of organizations because of the model's advanced cybersecurity capabilities. The company introduced Fable, a publicly available version of Mythos with additional safeguards, two months later.

The models — particularly Mythos — captivated Wall Street and government officials alike, and they helped to fuel Anthropic's meteoric rise this year. The company said in its amended prospectus that it generated $33.24 billion in revenue for the nine months ended Sept. 30, up roughly 1,200% from the same period last year.

Penn said Fable 5.2 is primarily intended for "power users or power use cases" since it excels at complex, open-ended technical work. Other models like Claude Opus 5.5, which Anthropic released in September, are better suited for daily tasks, so Penn said users should turn to Fable 5.2 when they need to tackle hard problems in engineering, mathematics and robotics.

"These are just net new things that are completely novel, and so they're frontier use cases," Penn said. "We expect that over time to maybe trickle to the Opus class, but really right now we do see a difference in terms of performance that we think those frontier use cases benefit from."

Anthropic said Fable 5.2 is generally available on Monday, and Mythos 5.2 is available to vetted organizations through the company's Life Sciences and Cyber Verification Programs.

WATCH: Anthropic’s breakneck rise: What to know before the IPO

What to Watch

AI outlook — possibilities, not facts

  • Anthropic's IPO will take place as soon as next week

    Likely · Within days

Open Questions

  • What is the exact date of the IPO?
  • Which stock exchange will Anthropic list on?
  • What specific safety evaluations were conducted on Fable 5.2 and Mythos 5.2?

Related Topics

This article was originally published by CNBC.

Related Stories

《TAIPEI TIMES》ezTravel reports data breach affecting over 690,000 users
Urgent·

《TAIPEI TIMES》ezTravel reports data breach affecting over 690,000 users

Taiwanese online travel agency Yiyou.com discovered on Saturday that its membership center had been attacked by an unusual external network, resulting in unauthorized access to the historical order contact information of more than 690,000 customers. Preliminary investigation showed that no credit card numbers were involved. The company has activated the security protection mechanism, entrusted a third-party network security team to investigate, and reported the case to the competent authorities and the police.

自由时报
2 min read
France accuses Iran of hacking the cooling systems in a number of major stores
BREAKING·

France accuses Iran of hacking the cooling systems in a number of major stores

France has accused Iran of being behind cyberattacks that raised the temperatures of refrigerators and freezers in hundreds of supermarkets in France and seven other European countries, causing food spoilage worth millions of euros. French Foreign Minister Jean-Noel Barrot announced European sanctions on the officials, and the Paris prosecutor opened an investigation against the Shahid Kawa unit linked to the Iranian Revolutionary Guard.

BBC عربي
2 min read
More on this topicanthropic