Anthropic debut supported by Amazon, Cloud 4 Opus and Sonnet AI

Anthropic, Amazon-With the Openai competitor, on Thursday launched its most powerful group Artificial intelligence Models still: Claude 4.

The company said two models called Claude Opus 4 and Claude Sonnet 4 determine the “new standard” when it comes to AI agents and “can analyze thousands of data sources, perform long tasks, write content and perform difficult actions”, after all.

Anthropic, founded formerly Top -1 trillion dollars in income for a decade.

Companies seemingly remaining behind from competitors.

According to Jared Kaplan, the Chief Scientific Director of the Anthropic Antrapic Director of the Anthropic Director of Anthropy, anthropic stopped investments in Chatbat, and instead focused on improving the ability of Claude to engage in complex tasks such as research and coding, even writing.

He also admitted that “the more difficult the task, the more risk it arises that the model will come out of the rails … And we are really focused on solving this so that people can delegate a lot of work in our models.”

“We train these models since last year and we really have been waiting for them,” Kaplan said in an interview. “I think these models are much stronger than the agents and as coders. It was definitely struggle internally just because some of the new infrastructure we used to workout these models … did it very down to the team wires in terms of doing everything and work.”

Kaplan added that as soon as they tried to force the model testers to turn off the new models because they needed to overestimate them, no one wanted to give up access.

Anthropic said that Claude Opus 4 was a “best coding model” and can run autonomously on a full corporate day – seven hours.

Both models can look for the Internet to complete the tasks on the username and alternate the reasoning and use tools, according to the anthrop. The company also stated that when you access local files, they could acquire and maintain “basic facts to maintain continuity and construction of silent knowledge over time.”

“I write a lot with Clood, and I think before Opus 4 and Sonnet 4 I mostly used models as a partner’s thinking partner, but still engaged in writing,” Mike Crger said in an interview with an anthropic product. “And they crossed this threshold, where most of my writings are actually … Opus basically, and now it is unknown from my writing.”

Kriger added: “I like that we are kind of pressing the border on both sides. Like one, it is a coding work and agency in general, and it makes many coding startups … But then we also push the border on how these models can learn and then be a really useful partner.”

In the first quarter Anthropic annual income reached $ 2 billion, the company confirmed Last weekmore than doubled at a speed of $ 1 billion in the previous period. In a recent interview with CNBC, Kate Jensen’s revenue said that the number of customers spends more than $ 100,000 annually with anthropic jumped up eight times A year ago.

Wall Street continues to pour money into AI startups as anthropic: the company received 2.5 billion dollars, a five-year rotation credit line Last week to strengthen its liquidity in the permanent and expensive AI-expensive competition.