Anthropic, supported by Amazon and the competition for Openai, launched the most powerful set of artificial intelligence models so far under the name Claude 4. The company stated that the two new models, Claude OPUS 4 and Clauds Sonnet 4, represent a “new standard” in the world of artificial intelligence agents, as “they can analyze thousands of data sources, Implementing long -term tasks, writing content in human quality, and performing complex works. Claude Sonnet 4 is a more efficient and lowest model, which is the most appropriate for public tasks, and comes as an alternative to Sonnet 3.7 model launched last February. Anthropic says Sonnet 4 provides “high programming and inference” and provides more accurate responses. The company adds that both models are 65% less likely to resort to tricks or brief ways to perform tasks compared to SonNet 3.7, which are better in retaining basic information for long -term tasks when Claud is provided with access to local files. One of the new advantages that was introduced to my model Claude 4 is the feature of “thinking summaries”, which simplifies the process of thinking in the model into easily understood visions. The “extended thinking” feature was also launched in its experimental version, which allows users to switch between thinking situations or the use of tools to improve the performance and accuracy of answers. Anthropic was founded by former officials of the Openai Research Department, and Claude’s conversation robot launched in March 2023. Since then, it has become part of the escalating arms race in the field of artificial intelligence between startups and technical giants, in a market that is expected to exceed trillion dollars of revenue within a decade. In various sectors, companies rush to integrate artificial intelligence robots and smart agents in their services for fear of defeating competitors. Anthropic has suspended its investments in conversation robots at the end of last year, and instead focused on improving Claud’s ability to perform complex tasks such as research and programming, and even writing full software rules. Anthropic also described the Claude OPUS 4 model as “the best programming model in the world”, noting that he can work independently for nearly a full working day; Any seven hours. Models can browse the Internet to accomplish tasks on behalf of the user, and to switch between patterns of thinking and the use of tools, according to the company. She added that, when providing them with access to local files, they can extract and preserve “basic facts to maintain continuity and build implicit knowledge over time.” The company also confirmed last week that its annual revenues amounted to 2 billion dollars in the first quarter of the year, more than twice the previous revenue rate of one billion dollars. Wall Street is still pumping money into emerging artificial intelligence companies such as anthropic; Last week, the company obtained a 2.5 billion dollar credit line for five years to enhance its financial load in light of the increasing and expensive competition in this field.
