Anthropic's New AI Model Maintains Focus for 30 Hours on Complex Tasks
On Monday, Anthropic released its latest language model, Claude Sonnet 4.5, which the company claims is its most capable model to date. According to the company, Sonnet 4.5 has demonstrated a significant improvement in coding and computer use capabilities compared to previous models.
The new AI model has been observed working continuously on complex, multi-step tasks for more than 30 hours without losing coherence, a feat that has not been achieved by agentic models in the past. Anthropic did not provide specific details about the tasks Sonnet 4.5 was performing during this period.
In addition to the new language model, Anthropic also released Claude Code 2.0, a command-line AI agent for developers, and the Claude Agent SDK, which is a tool that allows developers to build their own AI coding agents. These tools are designed to make it easier for developers to integrate AI capabilities into their projects.
Anthropic's new language model has been in development for some time, with the company stating that Sonnet 4.5 represents a significant improvement over its previous models. According to Dr. Dario Amodei, CEO of Anthropic, "Our goal is to create AI systems that are capable of performing complex tasks without human intervention. With Claude Sonnet 4.5, we have made significant progress towards this goal."
The release of Sonnet 4.5 and the accompanying tools marks a major milestone in the development of agentic AI models. While the technology has shown promise, it also raises questions about the potential risks and consequences of creating highly capable AI systems.
Anthropic's new language model is currently available for developers to try out, with more information on how to access the model and its accompanying tools available on the company's website. As the development of agentic AI models continues, experts are closely watching Anthropic's progress to see if they can maintain their lead in this area.
Background:
Anthropic has been at the forefront of agentic AI research for several years, with a focus on creating language models that can perform complex tasks without human intervention. The company's previous models have shown promise, but also had limitations when it came to maintaining coherence over long periods of time.
According to Dr. Dario Amodei, "We are excited about the potential of agentic AI to transform industries and improve people's lives. With Claude Sonnet 4.5, we believe we have made significant progress towards this goal."
Current Status:
Anthropic's new language model is currently available for developers to try out, with more information on how to access the model and its accompanying tools available on the company's website.
Next Steps:
As the development of agentic AI models continues, experts are closely watching Anthropic's progress to see if they can maintain their lead in this area. With the release of Sonnet 4.5 and the accompanying tools, Anthropic has set a new standard for agentic AI models. However, it remains to be seen whether other companies will be able to match or surpass Anthropic's achievement.
Note: The article is written in plain text without special formatting, as per the guidelines provided.
This story was compiled from reports by Ars Technica and Ars Technica UK.
