Claude API

Definition

Anthropic's service that lets programs send messages to Claude and get replies back, so developers can build Claude into their own apps.

More from How AI Models Work

• Agentic AI

  AI that works toward a goal over many steps on its own: it plans, takes actions with tools, checks the results, and keeps going until the job is done.

• Chain of Thought

  Having a model work through a problem step by step before giving its final answer, which often makes the answer more accurate.

• Context Window

  How much text Claude can keep in view at once, counting your messages, the files it has read, and its own replies.

• Few-Shot Prompting

  Giving a model a few worked examples of what you want before your real request, so it copies the pattern.

• Fine-Tuning

  Training an existing AI model further on a focused set of examples so it gets better at one particular job. A prompt shapes one conversation, but fine-tuning changes the model itself.

• Hallucination

  When an AI model states something made up as if it were fact, such as a quote, a number, or a file that doesn't exist.