Anthropic unveils latest AI model, aiming to extend its lead in coding intelligence

On Monday, Anthropic announced the release of its latest AI model, Claude Sonnet 4.5, claiming it to be the leading system for coding intelligence worldwide. This new version arrives just four months after its predecessor, Sonnet 4, highlighting the startup's swift pace in the evolving field of generative AI. According to Anthropic, Sonnet 4.5 achieves top-tier results on SWE-Bench Verified, a key benchmark for assessing software engineering performance. The company emphasized the model’s capability to drive significant business outcomes through autonomous computing, enhanced cybersecurity features, and the development of production-ready applications alongside context-aware AI agents. Anthropic's financial growth this year has been notably fueled by the coding functionalities of its models, particularly with a product known as Claude Code, which alone is projected to generate over $500 million in run-rate revenue, witnessing a tenfold increase in usage within just three months. The introduction of Sonnet 4.5 is seen as a strategic move to maintain its competitive edge in the AI coding landscape. The potential for automated and assisted software coding is significant, offering substantial productivity improvements and cost efficiencies. Sonnet 4.5 is designed to enhance code reliability, optimize refactoring decisions, and ensure production readiness, positioning it against competitors like Google’s Gemini, OpenAI’s GPT-5, and xAI’s Grok 4. From a business perspective, the implications of this new model are extensive. In the realm of cybersecurity, Sonnet 4.5 is reported to accelerate the detection and resolution of vulnerabilities. In the financial sector, it outperforms Anthropic's previous model, Opus 4.1, in various tasks including research, modeling, and forecasting. Most strikingly, the new model boasts the ability to function autonomously for up to 30 hours, a remarkable increase in endurance compared to Opus 4. To support developers, Anthropic is expanding its ecosystem with new tools that make Claude Code's capabilities more accessible. These enhancements include virtual machines, memory management, and context management tools, along with a native VS Code extension and improved terminal workflows. Additionally, the Claude Developer Platform now features a Claude Agent software development kit, offering developers advanced tools for creating tailored, context-aware AI agents.