Claude Sonnet 4.5 - Anthropic's latest AI programming model

What is Claude Sonnet 4.5?

Claude Sonnet 4.5 is Anthropic's latest and most powerful programming model. It excels in multiple areas including programming, computer operation, reasoning, and mathematics, topping the SWE-bench Verified test and capable of sustained work for over 30 hours. The model possesses powerful agent capabilities, automating complex tasks through APIs and tools. Claude Sonnet 4.5 features significant improvements in alignment and security, reducing malicious behavior and enhancing defenses. It supports various application scenarios, including code generation, knowledge-based question answering, and content creation. Claude Sonnet 4.5 is now fully available through the claude.ai website, app, and API (API pricing is the same as its predecessor, Claude Sonnet 4).

Claude Sonnet 4.5 introduces the "Imagine with Claude" feature, allowing users to interact with Claude via natural language commands, generating and adjusting software code and functionality in real time. All content is created instantly, without the need for pre-defined code. This feature is currently only available to Max subscribers and aims to accelerate software development, prototyping, and creative exploration, providing an efficient and flexible interactive experience.

Key features of Claude Sonnet 4.5

• Long-running intelligent agentsIt provides superior instruction following, tool selection, error correction, and advanced reasoning capabilities, suitable for customer-facing intelligent agents and complex AI workflows.
• Code generationSonnet 4.5 can complete the entire software development lifecycle tasks from initial planning to vulnerability fixing, maintenance and large-scale refactoring, and supports up to 64K output tokens, making it suitable for complex code generation and planning.
• Browser and computer operationThe model boasts superior computing power, reliably handling a wide range of browser-based tasks, from competitive analysis to procurement processes and customer onboarding, with more accurate and reliable computing capabilities than its predecessor.
• CybersecurityTeams using Sonnet 4.5 and Claude Code can deploy intelligent agents to automatically patch vulnerabilities to prevent exploitation, shifting from passive detection to proactive defense.
• Financial AnalysisThe model can handle all tasks from basic financial analysis to advanced predictive analytics, such as continuously monitoring global regulatory changes, proactively adjusting compliance systems, and enabling a shift from manual audit preparation to intelligent risk management.
• Business tasksThe model performs exceptionally well in creating and editing office documents such as slides, documents, and spreadsheets.
• ResearchSonnet 4.5 can search external and internal data sources to synthesize comprehensive insights in complex information environments.
• Content generation and analysisThe model excels in writing, understanding nuances and tone, generating more engaging content, and analyzing content at a deeper level.

Claude Sonnet 4.5 core upgrade

• Claude CodeClaude Sonnet 4.5 adds checkpoint functionality to Claude Code, which can save progress and instantly revert to the previous state. It also updates the terminal interface and releases a native VS Code extension to meet strong user demand and improve the development experience.
• Claude APIClaude Sonnet 4.5 adds context editing and memory tools to the Claude API, enabling agents to run for longer periods and handle more complex tasks, further enhancing the API's functionality and application scenarios.
• Claude applicationClaude Sonnet 4.5 integrates code execution and file creation capabilities (including spreadsheets, slideshows, and documents) directly into conversations within the Claude app and provides a Chrome extension for Max users who joined the waitlist last month, enhancing user convenience and efficiency in actual use.
• Claude Agent SDKClaude Sonnet 4.5 introduces the Claude Agent SDK, providing developers with the infrastructure and tools to build Claude Code, enabling them to build their own cutting-edge products.

Performance of Claude Sonnet 4.5

• SWE-bench Verified AssessmentClaude Sonnet 4.5 achieved industry-leading performance in the SWE-bench Verified evaluation. In practical applications, the model can maintain focus for over 30 hours in complex, multi-step tasks.
• OSWorld BenchmarksIn the OSWorld benchmark, Claude Sonnet 4.5 showed a significant improvement, achieving an accuracy of 61.4%, outperforming other models.
• Chrome browserWith the Claude for Chrome extension, Sonnet 4.5 can work directly in the browser, including browsing websites, filling out spreadsheets, and completing tasks.
• In a wide range of assessments, including reasoning and mathematicsClaude Sonnet 4.5 demonstrates improved capabilities in a wide range of evaluations, including reasoning and mathematics, and performs better in these areas than previous models.
• Domain-specific knowledge and reasoning abilityExperts in finance, law, medicine, and STEM fields have found that Sonnet 4.5 significantly improves knowledge and reasoning abilities in specific domains compared to older models (including Opus 4.1).

Application scenarios of Claude Sonnet 4.5

• Long-running intelligent agentsSuitable for complex tasks that require long-term operation, such as automated workflows, continuous monitoring, and multi-step task processing.
• Code generationIt spans the entire software development lifecycle, from planning to implementation, maintenance, and refactoring, supporting complex programming tasks.
• Browser and computer operationIt supports handling browser-based tasks such as automating web page operations, data scraping, and online workflow processing.
• CybersecurityDeploy intelligent physical capabilities to automatically detect and repair vulnerabilities, shifting from passive defense to proactive security protection.
• Financial AnalysisThe model can perform financial analysis from basic to advanced levels, including risk assessment, compliance monitoring, and predictive analytics.

Official report on Claude Sonnet 4.5

Free to claimClaude Sonnet 4.5 Official Model ReportScan the QR code to follow us on WeChat and reply:20250930

Who is Claude Sonnet 4.5 suitable for?

• Software developersThe model's powerful code generation and optimization capabilities help developers quickly implement complex programming tasks and improve development efficiency.
• Cybersecurity TeamThe model can automatically detect and fix vulnerabilities, helping teams shift from passive defense to proactive security protection.
• Financial AnalystThe model can handle financial analysis from basic to advanced levels, including risk assessment and compliance monitoring, facilitating intelligent risk management.
• Business personnelEfficiently create and edit office documents such as PowerPoint presentations, documents, and spreadsheets to improve daily office efficiency.
• ResearchersIntegrating internal and external data sources to generate comprehensive insights across complex information domains, supporting academic and business research.