Anthropic has announced two advanced artificial intelligence models, Claude Fable 5.2 and Claude Mythos 5.2, marking a significant development in its efforts to strengthen its position in the global AI market. The models were introduced on November 9, 2026, with improvements in coding, reasoning, cybersecurity, scientific research and engineering.
The launch comes at an important time for Anthropic as the company prepares for a potential initial public offering (IPO) and faces growing competition from major AI developers, including OpenAI and Google.
According to Anthropic, the new models are designed to handle complex technical challenges that previously required extensive human expertise and considerable development time.
Claude Fable 5.2 and Mythos 5.2: What Is New?
Anthropic describes Fable 5.2 and Mythos 5.2 as versions of the same underlying AI model, with different safeguards and access arrangements.
Claude Fable 5.2 includes stronger cybersecurity and biological safety protections, making it suitable for broader availability. Claude Mythos 5.2 is designed for specialised cybersecurity and biomedical research applications and is restricted to approved organisations.
The company says both models deliver outputs up to 30% faster than their predecessors while offering improved performance at a lower standard usage price.
Major Improvements in Coding and Reasoning
One of the most significant developments is the models’ ability to manage complex software engineering projects.
Anthropic reported that more than 90% of changes merged into its core codebase are now written by Claude, with human guidance.
In one experiment, an earlier version of Fable 5.2 helped migrate a critical production service from Python to Rust. The resulting service reportedly became more efficient and reliable, with startup time reduced by approximately 26 times.
Another experiment involved more than 1,000 AI agents working together for eight days to build a functional operating system from an empty folder. The project included core system functionality, drivers, file storage, a graphical interface and an installer.
These examples highlight Anthropic’s focus on AI systems capable of coordinating multi-step technical work rather than simply responding to individual prompts.
Major Advances in Cybersecurity
Claude Mythos 5.2 is being positioned as Anthropic’s most capable model for cybersecurity engineering.
The company says the model can identify software vulnerabilities, reproduce security issues and help developers fix them. In Anthropic’s internal CVE Rediscovery evaluation involving 1,829 known vulnerabilities, Mythos 5.2 reportedly identified 10.2% more vulnerabilities than Claude Opus 5.5 and 14.4% more than Mythos 5.1.
Anthropic also reported that an external security company found that one in six confirmed bugs identified by Mythos 5.2 had been missed by the other frontier models it tested.
Because advanced cybersecurity capabilities can also be misused, access to Mythos 5.2 is restricted through Anthropic’s Cyber Verification Program. Enterprise customers can also access related security capabilities through Claude Security.
AI Makes Progress in Scientific Research
Anthropic highlighted several research achievements involving the new models.
In theoretical physics, Fable 5.2 reportedly helped solve two previously open problems related to black holes and quantum gravity.
In mathematics, a coordinated team of approximately 1,000 Fable 5.2 agents formalised a proof of Fermat’s Last Theorem in the Lean programming language in around 30 hours, with minimal human intervention.
The company also described work involving protein structure prediction, hardware design and robotics. In one hardware experiment, the model helped design three electronic devices, including their circuit boards, enclosures and software.
These demonstrations suggest that AI systems are increasingly being used as research assistants and engineering collaborators across specialised fields.
Faster Performance and Updated Pricing
Anthropic says Fable 5.2 and Mythos 5.2 generate outputs up to 30% faster than their previous versions.
The standard API pricing announced for the models is:
| Model pricing | Cost |
|---|---|
| Input tokens | $8 per million |
| Output tokens | $40 per million |
| Fast mode input | $16 per million |
| Fast mode output | $80 per million |
The optional fast mode can increase token generation speed by up to 2.5 times, according to Anthropic.
Claude Fable 5.2 is available across Anthropic’s platforms and selected cloud services. Mythos 5.2 remains limited to eligible organisations and approved research programmes.
Anthropic’s Potential IPO Adds Commercial Significance
The launch also comes amid preparations for Anthropic’s proposed stock market listing.
Anthropic announced on October 26, 2026, that it had filed a registration statement with the US Securities and Exchange Commission for a proposed initial public offering of its Class A common stock.
The new model release could strengthen the company’s commercial positioning by demonstrating advances in enterprise AI, software development, cybersecurity and scientific research.
However, the timing of the launch does not establish that an IPO is imminent. The listing remains subject to the regulatory and market processes involved in a public offering.
What the Launch Means for the AI Industry
The introduction of Fable 5.2 and Mythos 5.2 reflects a broader shift in the AI industry towards systems that can complete complex, multi-stage tasks.
Competition is increasingly focused on coding quality, autonomous agents, scientific discovery, enterprise productivity, security and the cost of running advanced models.
For businesses, developers and researchers, more capable AI systems could reduce development time and make sophisticated technical tools more accessible. At the same time, stronger safeguards and responsible access controls will remain essential as AI capabilities expand.
Anthropic’s latest release adds another major development to the global AI race, while its proposed IPO could bring additional attention to the company’s growth, technology and long-term business strategy.
