What’s new with Claude 4? And why it’s becoming my favorite AI tool

Chatgpt is by far the most famous Chat IA service, but that does not automatically do the best. Google Gemini is constantly at the top of the rankings and does not count Claude.

If you are not familiar with Claude, it is the family of large language models manufactured by the company Ai Anthropic. And Claude has just received a huge upgrade in the form of Claude 4, the new model of Anthropic AI. Technically, Claude 4 is two models: Claude Opus 4 And Claude SONNET 4.

These new models bring a lot to the table with regard to the features of the AI, including improved reasoning and coding that competes with everything that is done by Chatgpt-Maker Openai. Here is an overview of what is new with Claude 4, and all the reasons why you could consider using Claude instead of Chatgpt.

Claude Opus 4 vs Claude Sonnet 4

Claude 4 is not in fact a model in large language – it is a family large -language models. Currently, there are two models in this family – Claude Opus 4 and Claude Sonnet 4, both that I used. Aigle eyes users may wonder if I have the right name, but indeed, Anthropic went from “Claude 3.7 Sonnet” to “Claude Sonnet 4.” I better like the new name, but of course, the change in name is a bit confusing.

So what is the difference between Claude Sonnet 4 and Claude Opus 4? In simple terms, Claude Sonnet 4 is the daily model you would use to ask basic questions, find information, etc. Claude Opus 4, on the other hand, is useful for advanced coding.

According to Anthropic, the two Claude 4 models are better than their predecessors significantly. To begin with, the company claims that they are 65% less likely to take shortcuts in certain agent workflows which could be sensitive to shortcuts and gaps. Not only that, but Claude Opus 4 also has well improved memory capacities, which allows him to better remember information in development workflows. For example, it can create memory files to refer without the need for a larger context window (the basic amount of data that a model can process both).

There is a compromise for the more advanced Claude Opus 4. Since this is a much wider model, it is much more expensive to use and, as such, it is not available for free users. Even those who pay for the subscription of Claude Pro of $ 20 per month can reach limits after using Claude Opus 4. Those who subscribe to the maximum account of $ 100 per month will however have much higher limits.

Why I like Claude

There are a few things that I really like in Claude, although it is important to note that there are still a lot of features that you will not get in Claude against another AI service. The greatest thing I love in Claude is that it is a little more … like man.

That is to say, everything, from design to the types of answers that I get, seems a little more personal and a little more natural. This is true for Claude Sonnet 4 and Claude Opus 4. There are small touches, like the small “thought” icon, to more important differences, like the type of text it generates. Part of this has to do with the fact that you can adapt each cat with Claude with preset and personalized “styles”, which dictate how Claude will react. The predefined options include “concise” and “explanatory”, but you can also create your own styles.

Claude against competition

While AI’s services continue to improve at dizzying speed, all AI services do not offer all functionality. That is to say, there are many features that are not yet available to Claude who are Available in chatgpt, for example.

For example, Claude cannot generate images, and Anthropic has just started to deploy a vocal chat service in Claude. Still in beta, Claude’s vocal cat does not seem as natural as the textual version. These are perhaps table issues for you-and if you count a lot on this type of functionality, then Claude may not be for you, or it may not be the only AI service you need.

But Claude also gets a lot RIGHT. For example, Claude has a “deep research” type functionality (simply called research), which analyzes many sources to find information. It also has an “extensive reflection” functionality, which is essentially a characteristic of reasoning in which it will think about its response longer before responding, which makes it possible to respond to more precise responses.

Although Claude has become much better for tasks like simple research and daily use, its best use case is always coding. To be clear, I am not more than a basic coder, but according to all accounts, Claude is better than any other coding service. Anthropic, the manufacturer of Claude, even created an autonomous service called Claude Code, which is a specially designed coding assistant which is powered by Claude Opus 4.

The good news is that you do not necessarily have to pay to try Claude, but if you do not do, you will not be able to use its most advanced features. Instead, all you have to do is create an account and try it, which you can do before paying for an idea of ​​the service and if that suits you.

Disclosure: Ziff Davis, Mashable’s parent company, in April, filed a complaint against Openai, alleging that it has violated Ziff Davis Copyrights in the training and exploitation of its AI systems.