Which Claude Model Should You Use? Anthropic’s Recommendations by Task
Addy Osmani from Anthropic has published a guide on which model for a task.
Opus 5.5 (medium)which has better balance of everything and it is it recommended for coding and general.Sonnet 5.5when you know exactly what needs to be done with clear specHaiku 5.5for routing, summarization and other minor tasksFable 5.2for deep research and hard/complex coding task
