Anthropic's Newest AI Model Opus 5 Is Now Available
However, it fell somewhat short of Fable 5 in task categories such as answering legal questions and performing multidisciplinary reasoning without additional tools. It's also "substantially behind" Mythos 5, Anthropic's closed model only shared with selected partners, at exploiting cybersecurity vulnerabilities. The behavior suite found the newest model exhibited lower rates of deceptive behavior than previous models, and it proved harder to trick into misuse.
- ▪However, it fell somewhat short of Fable 5 in task categories such as answering legal questions and performing multidisciplinary reasoning without additional tools.
- ▪It's also "substantially behind" Mythos 5, Anthropic's closed model only shared with selected partners, at exploiting cybersecurity vulnerabilities.
- ▪The behavior suite found the newest model exhibited lower rates of deceptive behavior than previous models, and it proved harder to trick into misuse.
2 outlets in our directory ran this story, first to last over 19 hours. All of the coverage we found sits in one bucket: centre. That one-sidedness is itself worth noticing.
PCMag files mainly under tech. We currently carry 254 of its stories.
Opening excerpt (first ~120 words) tap to expand
Anthropic's newest model, Claude Opus 5, is now available to the public.Opus 5 comes close to or even exceeds its predecessor Fable 5's performance in most areas, while costing roughly half as much per token, according to the AI giant's own benchmarking figures.The new model outperformed Fable 5 in tasks like knowledge work, novel problem solving, and agentic search. However, it fell somewhat short of Fable 5 in task categories such as answering legal questions and performing multidisciplinary reasoning without additional tools. It's also "substantially behind" Mythos 5, Anthropic's closed model only shared with selected partners, at exploiting cybersecurity vulnerabilities.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at PCMag.
Discussion
0 commentsMore from PCMag
