The default mode for most development work. Uses 3 models with medium thinking effort for proper planning and execution. Ideal for feature implementation, debugging, refactoring, and analysis. Claude Sonnet 4.5 consensus.
Try NowChanges across multiple files
Implementing new features or components
Debugging issues or fixing bugs
Refactoring code structure
Writing or updating tests
Questions requiring code or documentation reading
Task involves any uncertainty or decision-making
Needs understanding of code context
Requires any form of analysis or reasoning
3 models
Medium
2,048 tokens
Claude Sonnet 4.5
This mode uses 3 AI models working in parallel to analyze your request. Their responses are then synthesized by Claude Sonnet 4.5 to produce the final answer. This approach improves accuracy by leveraging multiple perspectives and cross-validating conclusions.