- Anthropic made his debut two new models Claude AI named Claude Opus 4 and Claude Sonnet 4
- Claude Opus 4 claims to be the best coding AI in the world
- Claude Sonnet 4 is a smaller and rationalized model with major upgrades of the Sonnet 3.7 version.
Anthropic unveiled Claude 4, the latest generation of its AI models. The company boasts that the Claude Opus 4 models and Claude Sonnet 4 are at the top of the game for AI assistants with unrivaled coding skills and the ability to operate independently for long periods.
Claude Sonnet 4 is the smaller model, but it is always a major upgrade of the energy of the Old Sonnet 3.7. Anthropic claims that Sonnet 4 is much better to follow the instructions and coding. It was even adopted by Github to feed a new co -pilot coding agent. It is probably much more used simply because it is the default model on the free level of the Claude chatbot.
Claude Opus 4 is the flagship model for anthropic and supposedly the best coding AI. He can also manage supported tasks of several hours, dividing them into thousands of steps to be accomplished. Opus 4 also includes the “extended reflection” function anthropic on previous models. The extended reflection allows the model to stop in the middle of responding to an prompt and using search engines and other tools until it had more data and can take up the moment when it had stopped.
This means much more than longer answers. Developers can train Opus 4 to use all kinds of third -party tools. Opus 4 can even play video games quite well, with Anthropic showing how AI works during a Pokémon Red game when accessing the file and permission to create its own navigation guide.
Claude 4 Power
The two Claude 4 models have improved features centered on the use and memory of the tools. Opus 4 and Sonnet 4 can use tools in parallel and switch between reasoning and research. And their memory system can record and extract key facts over time when they have given access to external files. You will not have to re -explain what you want with each third prompt.
To ensure that AI does what you want, but do not overwhelm yourself with every detail, the models of Claude 4 also offer what it calls “summaries of reflection”. Instead of a wall of text detailing each thousands of measures potentially taken to complete an prompt, Claude uses a smaller secondary AI model to condense the thought train in something digestible.
A secondary advantage of how new models work is that they are less likely to cheat to save time and treatment power. Anthropic said they had reduced shortcut’s research behavior in tasks that try AI to simulate their path to a solution (or just invent something).
Overview? Anthropic clearly takes the lead in the usefulness of the AI, in particular in coding and tasks independent of the coding and independent. Chatgpt and Google Gemini have larger user bases, but Anthropic has the means to attract at least some Chatbot AI users in Claude. With Sonnet 4 available for free users and Opus 4 joined Claude Pro, Max, Team and Enterprise Plans, Anthropic tries to please AI fans both favorable to a budget and premium.
