API
An interface that lets one program talk to another — in AI, usually the way developers send prompts to a model and receive responses programmatically instead of through a chat UI.
An API is a defined set of rules by which one program can call another program's functionality. For AI, the API is how developers integrate a model into their own products — send an HTTP request with a prompt, receive a response with the model's output.
Anthropic's Claude API, OpenAI's API, Google's Gemini API, Mistral's API — these are the underlying interfaces that consumer products like Claude.ai, ChatGPT, and LADLE build on top of. When you chat with LADLE, we forward your message to Anthropic's Claude API, get the streamed response, and render it in our UI.
The distinction matters because API access is generally cheaper per-query but requires you to build the UI and handle the plumbing yourself. Consumer products (chat UIs with subscription pricing) trade that convenience against monthly cost. For a developer building an AI feature into their own product, calling the API directly is usually the right choice; for someone who wants to chat, a consumer product is.