Skip to main content

Glossary / Model routing

Glossary

Model routing

In short

Sending each task, or each part of a task, to the model best suited for it, rather than using one model for everything.

Different models have different strengths and prices. Routing means matching the work to the model: a cheap, fast model for routine steps, a strong reasoning model for the hard part, a model with a long context window for big documents.

Done well, routing gets you most of the quality of the best model at a fraction of the cost, because you are not paying premium rates for work that a smaller model handles fine.

In LLMWeave

Inside a weave, work is matched to the right model rather than forced through a single one. Cheaper models handle the routine passes, and stronger models are reserved for the steps that actually need them.

Related terms

Try multi-model on your task

One prompt, several models, one answer. Free to start, no card.

Get started