Ahropic unveiled its latest artificial ielligence model, Claude 3.7 Sonnet last night. This advanced model is designed to provide more accurate and thoughtful answers to user questions. The promine feature of Claude 3.7 SONNET is its “hybrid reasoning” feature, which allows users to provide both urge and deeper and more thoughtful responses.
According to Ahropic, Claude 3.7 SONNET is the first model of artificial ielligence to use this type of hybrid reasoning. Users can choose whether they wa to use the model reasoning. Activating this feature allows Claude 3.7 Sonnet to think about questions for a shorter or longer period of time and provide higher quality answers. In terms of pricing, Claude 3.7 SONNET is more expensive than some competing models such as O3-Mini than Openai and Deep Sik R1. However, ahropic emphasizes that Claude 3.7 SONNET is a hybrid model and offers more capabilities than purely reasonable models such as O3-MINI and R1.
In order to enhance the transparency and trust of users, the ahropic provides users with the Claude 3.7 Sonnet thinking process through a draft draft page. This feature allows users to see how the model is reasonable and reached. Claude 3.7 SONNET has been optimized for a variety of applications, including complex coding problems and performing tasks related to smart ages. Developers can corol the “budget” allocated for model thinking through the ahropic API, thereby creating a balance between speed, cost and quality of responses.
In the SWE-Bench test for evaluating coding tasks, the model gained 62.3 % accuracy and earned 81.2 % in the Tau-Bench test to measure ieraction with simulated users. In addition to advanced capabilities, Claude 3.7 SONNET has also been improved in safety. Ahropics claims that this model refuses to answer harmless questions less than previous models and makes a more accurate distinction between harmful and safe requests.
At the same time as the Claude 3.7 SONNET iroduced, the ahropic has also launched a new tool called Claude Code. This tool allows developers to perform coding tasks directly through the terminal and using a cloud. The supply of Claude 3.7 SONNET comes in a situation where competition in the field of artificial ielligence has increased sharply and various laboratories are rapidly iroducing new models. The ahropic, which has always been known for a cautious and safety -orieed approach, has shown that it is a pioneer in the competition and provides users with advanced and useful artificial ielligence models.
Source: The Verge
