🧠 AI Models / /via cnbc.com / updated Oct 10, 2026

Anthropic Launches Cheaper Sonnet 5.5 AI Model With Stronger Cyber Safeguards

Anthropic launched Sonnet 5.5, a faster, lower-cost AI model designed for coding, scoped tasks and polished business documents. It is the company’s second model release since CEO Dario Amodei urged AI companies to slow the pace of progress on their most advanced systems. The model costs $2 per million input tokens and $10 per million output tokens while introducing cyber safeguards previously used for Anthropic’s most capable models.

#Anthropic#AmazonWebServices#GoogleCloud#MicrosoftAzure
~/ AI Models/ Anthropic Launches Cheaper Sonnet 5.5 AI Model ...

Anthropic on Monday released Sonnet 5.5, a faster and lower-cost artificial intelligence model aimed at customers who need reliable execution without the higher cost of the company’s most capable offering.

The company said Sonnet 5.5 improves on its predecessor at coding, completing scoped tasks and creating polished documents, slides and spreadsheets. Theo Chu, a research product manager at Anthropic, described Sonnet as a fit for routine work that requires execution but not the deeper judgment associated with the company’s Opus models.

Sonnet 5.5 arrives less than a week after Anthropic introduced Opus 5.5, its more expensive model. Anthropic also said Haiku 5.5, the least expensive model in the suite, is coming soon.

The release is Anthropic’s second since CEO Dario Amodei called on AI companies to slow the pace at which they improve their most advanced models. That appeal came as researchers increasingly warned about the possibility of catastrophic harm from artificial intelligence.

Anthropic said Sonnet 5.5 does not advance the frontier of its models’ capabilities. As a result, the company said most of its alignment testing focused on a targeted set of risks relevant to models across capability levels.

At the same time, Anthropic described Sonnet 5.5’s cybersecurity capabilities as a large improvement over its predecessor. It is the first Sonnet model to launch with fallbacks and cyber safeguards similar to those Anthropic developed for its most capable models.

Sonnet 5.5 is priced at $2 per million input tokens and $10 per million output tokens, half the price of Opus 5.5. Anthropic said the model also uses fewer tokens per task than its previous version, which can further reduce the cost of running it.

Why this matters

The launch shows how AI companies are separating everyday productivity improvements from advances at the frontier of model capabilities. For businesses, a model positioned around routine coding and knowledge work could make broader deployment more economical, while the stronger cyber safeguards reflect the growing pressure to pair capability gains with tighter controls.

Sonnet 5.5 was available immediately across Amazon Web Services, Google Cloud and Microsoft Azure. Anthropic’s next planned step is the arrival of Haiku 5.5, which will complete the company’s latest model lineup with a lower-cost offering.

source cnbc.com →
share
𝕏 FB
← cd ../news