Model router
Model router
Version: 2025-11-18
Microsoft•Last updated November 2025
Model router is a deployable AI model that is trained to select the most suitable large language model (LLM) for a given prompt.
Multipurpose
Multimodal
Welcome to the 2025-11-18 version of model router. Model Router dynamically selects the optimal large language model (LLM) for a specific query or task in real time. By evaluating factors like query complexity, cost, and performance, it efficiently routes requests to the most suitable model, ensuring high quality results while minimizing costs. This version adds several new capabilities:
  1. Support Global Standard and Data Zone Standard deployments.
  2. Adds support for new models: grok-4, grok-4-fast-reasoning, DeepSeek-V3.1, gpt-oss-120b, Llama-4-Maverick-17B-128E-Instruct-FP8, gpt-4o, gpt-4o-mini, claude-haiku-4-5, claude-sonnet-4-5 and claude-opus-4-1.
  3. Support for agentic scenarios including tools so you can now use it in the Foundry Agent service.
  4. Quick deploy or Custom deploy with routing mode and model subset selections.
  5. Routing mode: Optimize the routing logic for your needs. Supported options: Quality, Cost, Balanced (default).
  6. Model subs
notes.md
evaluation.md
Model Specifications
Context Length1048576
LicenseCustom
Training DataNovember 2025
Last UpdatedNovember 2025
Input TypeText,Image
Output TypeText
ProviderMicrosoft
Languages1 Language
Quick Start
Sign in to get startedView DocumentationAzure Support