Claude

Anthropic's family of large language models — AI assistants trained with a Constitutional AI approach emphasizing safety, helpfulness, and harmlessness.

Constitutional AI assistant: Claude is Anthropic's AI assistant, trained using Constitutional AI — a method in which the model is trained to evaluate its own outputs against a set of principles (a 'constitution') and self-revise. The approach attempts to produce a helpful, harmless, and honest AI assistant without relying solely on human feedback labeling. Claude's distinctive emphasis on nuanced, thoughtful responses and its 'character' (intellectual curiosity, care, directness) reflects Anthropic's design philosophy. Claude models have been deployed in various sizes (Haiku, Sonnet, Opus) and have been particularly noted for their performance on reasoning and writing tasks.