Anthropic Launches Claude Fable 5.2 and Mythos 5.2 AI Models Ahead of Planned IPO
Artificial intelligence developer Anthropic released two advanced foundation models designated as Claude Fable 5.2 and Claude Mythos 5.2 on Monday. The technological release arrives as the artificial intelligence venture prepares to launch an initial public offering during the week of November 16. Both software offerings share the same underlying model but carry different safeguards tailored for distinct user groups. Claude Fable 5.2 is accessible to the broader commercial public with extensive safeguards restricting dangerous biological and cybersecurity activities. The specialized Mythos 5.2 model operates with reduced system filters and is restricted exclusively to vetted institutions across cybersecurity and life sciences.
Technical Capabilities and Benchmarking Performance
Company documentation indicates that both updated models generate computational tokens up to 30% faster than prior 5.1 versions. Standard API access pricing is set at $8 per million input tokens and $40 per million output tokens. An optional accelerated processing tier delivering speeds up to 2.5 times faster is offered at $16 per million input tokens and $80 per million output tokens. Corporate researchers reported that 1,000 cooperating Fable 5.2 autonomous agents generated a verified proof of Fermat's Last Theorem within 30 hours. In comparative software engineering benchmarks, Fable 5.2 recorded a 69.9% score on FrontierSWE v2, surpassing competitive frontier models.
Vulnerability Detection and Hardware Collaboration
On specialized vulnerability rediscovery assessments, Mythos 5.2 identified 10.2% more security weaknesses than the established Opus 5.5 framework. In hardware prototyping experiments with engineering firm Diode, the software successfully designed circuit boards for three physical electronic consumer devices. The hardware designs encompassed a handheld virtual pet, an integrated USB conference speakerphone, and a smart home voice assistant hub. Despite the advanced capabilities of the new releases, company engineers still recommend that clients utilize Opus 5.5 for routine software coding. Outside evaluators including METR, Frontier Design and Britain's AI Security Institute checked the model before release, according to Anthropic.
The original iterations of Fable and Mythos faced regulatory scrutiny after the United States Commerce Department required access restrictions following their June debut. Anthropic's revenue reached $17 billion in the third quarter, up from $1.4 billion a year earlier, according to a securities filing. Company documentation notes that automated safety classifiers block roughly 55% fewer legitimate requests compared to earlier filtering architectures. The dual-model deployment structure reflects an emerging industry strategy balancing broad public commercialization with tightly controlled advanced capabilities. Anthropic's initial formalization of Fermat's Last Theorem using an earlier research model took 11 days. Claude Mythos 5.2 is restricted exclusively to organizations approved through Anthropic's Cyber Verification or Life Sciences Verification programs, which require institutional compliance screening before lifting standard safety classifiers, according to AlphaCorp AI.