Anthropic has released Claude Sonnet 4.5, calling it the world’s best AI for real-world agents, programming, and computer-based tasks. It outperforms previous models in coding, logical reasoning, and handling long-context scenarios. The new version can now run autonomously for up to 30 hours – a major leap from the 7-hour limit of Claude Opus 4.
Alongside the model, Claude Code is getting new features like checkpoints and the ability to create spreadsheets, slides, and documents directly within the chat. Claude’s Chrome extension is also being rolled out more widely. Early users from companies like Canva and iGent AI praise Sonnet 4.5 for its significantly improved performance, speed, and flexibility when handling complex tasks.