Introduction of Claude Sonnet 4.5 by Anthropics
Anthropics has recently unveiled a new AI model, Claude Sonnet 4.5, marking a significant milestone in artificial intelligence technology. This model brings enhanced safety features and performance, reportedly surpassing both its predecessor and other leading models like OpenAI's GPT-5 and Google’s Gemini 2.5 Pro.
Innovations and Advantages of Claude Sonnet 4.5
Claude Sonnet 4.5 has undergone rigorous testing, scoring an impressive 82 percent on the SWE-bench Verified, which assesses real-world coding abilities. Noteworthy are the model’s sophisticated safety protocols designed to prevent deceptive functionalities and prompt injection attacks. These features are part of Anthropics' AI Safety Level 3 framework, aimed at ensuring the AI does not bypass security mechanisms or leak private information.
Applications and Cost-Efficiency
Mike Krieger, Chief Product Officer at Anthropics, highlighted the practical applications of Claude Sonnet 4.5. The AI is capable of handling continuous coding tasks for up to 30 hours, reflecting not only its endurance but also its cost-effectiveness, with pricing at $3 per million input tokens and $15 per million output tokens. Claude Sonnet 4.5 is available through Anthropics' API and chatbot interfaces since its launch on Monday.
Broader Impact and Market Positioning
By focusing on reliability and ethical standards, Anthropics is positioning Claude Sonnet 4.5 as a reliable choice for sectors like banking, healthcare, and government contracting. These sectors often require dependable technology solutions and are usually cautious in adopting new technologies. Anthropics' commitment to ethical guidelines and user privacy is also evident as it promotes responsible usage to ensure positive and respectful interactions for all stakeholders.
