Anthropic's Claude Fable 5.2 Beats GPT-6 Cyber in Multi-Stage Cyberattack Benchmark
Anthropic on Monday released Claude Fable 5.2 and Claude Mythos 5.2 models, which it said exhibit improvements in coding, reasoning, and vision. During pre-release training, the large language models (LLMs) helped its engineers "complete otherwise-infeasible…
Adjust this page to suit you. Your choices are saved in this browser.
Cookies. Chipokia uses one cookie, and only if you allow it: a random ID so your up and down votes stay yours. No analytics, no advertising, no tracking of any kind. Privacy & Cookie Policy.
