Anthropic has released Claude Opus 4.1, an upgraded version of their AI model with improved performance on coding tasks, achieving 74.5% on SWE-bench Verified. The model shows significant improvements in multi-file code refactoring, debugging precision, and agentic search capabilities. It's available to paid users through Claude's interface, API, Amazon Bedrock, and Google Cloud Vertex AI at the same pricing as Opus 4. Companies like GitHub, Rakuten Group, and Windsurf report notable performance gains, particularly in code accuracy and debugging tasks.