# Semantic Model Routing

> Semantic model routing reads each AI request and sends it to the model that handles it best for the lowest cost: simple, mechanical work to fast cheap models, hard reasoning to frontier models. It replaces the default of sending everything to the most expensive model in the building.

- Part of: [The Sentry AI Transformation Glossary](https://sentrysolutions.ai/glossary)
- Canonical: https://sentrysolutions.ai/glossary/semantic-model-routing

Most AI estates run on defaults: whichever model made the demo impressive handles everything forever after. The result is frontier prices for commodity work, because most enterprise traffic is classification, extraction and reformatting that a small model handles identically.

Routing makes the model choice a per-request decision owned by infrastructure rather than habit. On estates we run it is usually the single largest cost saving, and it works in both directions: it also stops cheap models grinding and retrying on work above their weight.

## Related terms

- [Conversation-Level Cost Tracking](https://sentrysolutions.ai/glossary/conversation-level-cost-tracking)
- [Agent Management Platform](https://sentrysolutions.ai/glossary/agent-management-platform)