A few days ago, Anthropic announced a new public language model of their generative artificial intelligence chatbot, Claude 3.7 Sonnet. It is a hybrid model that combines a regular model and a model that can think for itself, including more sophisticated reasoning.
It also took DeepSeek-R1 as inspiration, where users can see the process of Claude 3.7 Sonnet thinking. Through this new model, Claude Code was also introduced, which is a model that understands coding work in depth. With the era of AI models becoming AI agents to represent users in performing a task, this new Claude model has the ability to perform tasks and understand them with an accuracy rate of up to 81.2%.
Like other reasoning and thinking AI models, Claude 3.7 Sonnet is also great at mathematical calculations, solving physics problems, and performing more complex tasks such as research and analyzing reports. At the same time, Claude also ensures that their model is more focused on real-time tasks for enterprise and business industry use.
Claude 3.7 Sonnet is accessible to all free users, Pro, Team, and Enterprise plan subscribers – but features for deeper thinking are limited to paid plans.