Anthropic releases Claude Sonnet 4.5 in latest bid for AI agents and coding supremacy

Anthropic releases Claude Sonnet 4.5 in latest bid for AI agents and coding supremacy

Anthropic has introduced its new AI model, Claude Sonnet 4.5, which operated independently for 30 hours to develop a chat application similar to Slack or Teams. This model produced approximately 11,000 lines of code and halted only upon task completion. This marks a significant advancement compared to its predecessor, Opus 4, which only managed seven hours of autonomous operation.

The launch of Claude Sonnet 4.5 is part of Anthropic’s strategy to strengthen its position in AI agents and coding markets. The company describes this model as “the best model in the world for real-world agents, coding, and computer use,” highlighting its ability to excel in fields such as cybersecurity, financial services, and research. Clients, including Canva, reported that the model efficiently handled complex tasks, including engineering and product features.

Major companies like Anthropic, OpenAI, and Google continually update their AI technologies to cater to both consumer and enterprise needs. For instance, OpenAI recently launched Pulse, a new feature for ChatGPT aimed at enhancing users’ morning routines. Anthropic’s updates include support for developers to create their own AI agents using the new model.

The company has paired Claude Sonnet 4.5 with additional resources like virtual machines and multi-agent support to facilitate these tasks. Dianne Penn, Anthropic’s head of product management, noted that improvements in browser navigation and computer usage capabilities were unexpected, with the new model being over three times more efficient compared to earlier versions.

Scott White, product lead for Claude.ai, mentioned that the model operates at an advanced level, capable of managing calendar schedules and generating insights from data dashboards. While neither White nor Penn had experimented with vibe-coding during an interview, Penn shared that employing Claude Sonnet 4.5 has been beneficial in sourcing potential new team members through targeted web searches.

Source: https://www.theverge.com/ai-artificial-intelligence/787524/anthropic-releases-claude-sonnet-4-5-in-latest-bid-for-ai-agents-and-coding-supremacy

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top