Microsoft Foundry Model Router: one deployment that picks the best LLM per request in real time
lee_stott · x · 2026-09-04
Ahead of his Agentcon London talk, Lee Stott shared docs on Microsoft Foundry's Model Router:
- A trained routing model that sends each prompt to the most suitable underlying LLM in real time
- Deploys like any Foundry model—one deployment balancing performance, cost, and latency
- Works as a drop-in deployment or an optimization layer, replacing manual hill-climbing model selection
- Microsoft still recommends evaluating against your baseline to confirm managed routing helps your workload
More from coding & agent
- HKUST-GZ and Tencent open-source VibeWorlding, an agent that builds interactive 3D worlds from natural language — jiqizhixin · 2026-09-04
- Purdue's AutoTraceGT automates grounded theory coding to analyze agent behavior at scale — Purdue · 2026-09-04
- WebTerm Learn offers free browser-based terminal simulator to learn Git, CLI and Vim in 129 lessons — 4310sy · 2026-09-04
- Stalkr lands first customers in 10 minutes: social listening with AI-sorted mentions and MCP support — marclou · 2026-09-04
- Using MCP for CRM data cleanup: auditing 14k stale HubSpot accounts instead of prospecting — snowingbol · 2026-09-04
- GPT-6 Astra reviews: half the per-task cost, but CoT monitoring breaks down — vista8 · 2026-09-04