- Anthrope has presented two new Claude Ai models called Claude Opus 4 and Claude Sonnet 4
- Claude Opus 4 claims to be the best coding the world
- Claude Sonnet 4 is a smaller and optimized model with important updates of the Sonnet 3.7 version.
Anthrope has presented Claude 4, the last generation of his AI models. The company boasts that the new Claude Opus 4 models and Claude Sonnet 4 are at the top of the game for AI attendees with unique coding skills and the ability to function independently for long periods of time.
Claude Sonnet 4 is the smallest model, but it is still an important update in the power of Sonnet 3.7 above. Anthrope says that Sonnet 4 is much better to follow the instructions and coding. It has even been adopted by Github to feed a new Copilot coding agent. It is likely that it will be much more widely used because it is the predetermined model at the free level for the Claude Chatbot.
Claude Opus 4 is the flagship model for anthropic and supposedly the best AI of coding. You can also handle sustained tasks of several hours, dividing them into thousands of steps to fulfill. Opus 4 also includes the anthropic characteristic of “extended thought” proven in previous models. Extended thinking allows the model to pause in the middle of responding to a message and use search engines and other tools until it has more data and can resume right where it left it.
That means much more than longer responses. Developers can train Opus 4 to use all types of third -party tools. Opus 4 can even play video games quite well, with Anthrope showing how AI works during a Pokémon Red game when you are given file access and permission to build your own navigation guide.
Claude 4 Power
Both Claude 4 models have improved characteristics focused on the use of tool and memory. Opus 4 and Sonnet 4 can use parallel tools and change between reasoning and search. And your memory system can save and extract key facts over time when accessing external files is provided. You will not have to explain what you want in each third message.
To ensure that AI is doing what you want, but do not overwhelm it in every detail, Claude 4 models also offer what it calls “thought summaries.” Instead of a text wall detailing each of the thousands of steps taken to complete a notice, Claude uses a smaller secondary the AI model to condense the train of thought in something digestible.
A secondary benefit of the way in which the new models work is that it is less likely that Trampe to save time and processing power. Anthrope said that they have reduced the search for shortcuts in the tasks that tempt Ais to pretend their way to a solution (or simply invent something).
The biggest image? Anthrope is clearly shooting for leadership in the usefulness of AI, particularly in agents and independent coding and tasks. Chatgpt and Google Gemini have larger users bases, but Anthrope has the means to attract at least some AI Chatbot users to Claude. With Sonnet 4 available for free users and Opus 4 included in Claude Pro, Max, Team and Enterprise Plans, Anthrope is trying to attract both the fans of the budgets and premium.
