routing (LLM)
A workflow that classifies each input and sends it to a specialised prompt, tool or model.
In more detail
Routing lets each path be tuned on its own and sends easy requests to cheaper models. Give the router a way out, such as a fallback label or a human, for inputs it cannot classify confidently.
Related terms
- workflow (LLM) — A system in which your code fixes the sequence of model calls and tools, as opposed to an agent that chooses its own steps.
- prompt chaining — A workflow that splits a task into fixed steps, each a model call that feeds the next, often with a check in code between them.
- model router — A service that sends each request to one of many models or providers behind a single API, often with automatic fallbacks.
Taught in
- Choosing A Model — AI Engineering: LLM Apps, RAG & Agents
- The Request And The Response — AI Engineering: LLM Apps, RAG & Agents
- Workflows Before Agents — AI Engineering: LLM Apps, RAG & Agents
- Guardrails, Privacy And Governance — AI Engineering: LLM Apps, RAG & Agents