Unveiled today, Claude Sonnet 4.5’s ability to function independently for 30 hours is a notable advancement, particularly when compared to the company’s previous model, Opus 4, which gained attention for operating autonomously for seven hours back in May.

This progress marks a significant milestone in Anthropic’s efforts to dominate the AI agent and AI coding sectors. Describing Claude Sonnet 4.5 as “the world’s leading model for real-world agents, coding, and computer operations,” Anthropic highlights its superiority in managing computer tasks, especially in areas such as cybersecurity, financial services, and research. Canva, a beta-tester of the model, noted its effectiveness in handling “complex, long-context tasks” across various applications, including coding and research.

Competing companies like Anthropic, OpenAI, and Google continue to roll out updates and features designed to enhance both consumer and enterprise functionalities, ranging from research and scheduling to coding support and data analysis. The race for user engagement is intensifying with frequent developments. Recently, OpenAI introduced Pulse, a new ChatGPT feature tailored for daily routine and relevant topic exploration.

Anthropic also said the new model would be paired with other updates to help developers code their own AI agents.

Anthropic announced that alongside the model’s launch, they are providing resources like virtual machines and tools for memory and context management, which allow developers to create innovative agents using Claude Code’s foundational elements.

Dianne Penn, a product management leader at Anthropic, expressed her surprise at the model’s significant advancements in computer usage. Claude Sonnet 4.5 demonstrates more than triple the efficiency in web browsing and computer operation compared to Anthropic’s technology from the previous year. Penn mentioned that feedback from early-access clients like GitHub and Cursor has been instrumental during their recent development efforts.

Scott White, product lead for Claude.ai, told The Verge that the new model operates at “chief-of-staff level” and can find availability between multiple peoples’ calendars and schedule a meeting, look at a data dashboard and pull together insights, write status updates based on one-on-one meetings with his direct reports, and more.

Neither White nor Penn had yet tried vibe-coding with the new model when The Verge spoke to them. But Penn said she uses Claude Sonnet 4.5 for hiring potential new team members at Anthropic.

“It’s been actually really helpful to have a continuous running prompt that I use of, ‘Do a deep web search, come up with like these parameters for profiles to source for certain types of roles on my team,’” Penn said. “That’s been really, really helpful. And I’ve seen the Sonnet 4.5 just do even better than in the past, on the quality and the depth of the searches and actually generating a spreadsheet with LinkedIn profiles so then I can email them.”