Back to Courses
Course

Ray Serve for Online AI

Build Serve applications with deployment graphs, request routing, autoscaling, and LLM-ready controls.

Ray Serve for Online AI thumbnail

About this course

4 lessons
2 modules
2 hours
Mira Kapoor
Tags
Ray ServeModel servingLLM routing
Online Serving
Start course

1.Serve deployments

Compose model applications from independently scalable deployments.

  • Serve deployment grapharticle - 13 min
  • Deployment graph checkquiz - 6 min

2.LLM routing

Route requests across model sizes, adapters, and latency targets.

  • LLM routing with Ray Servearticle - 13 min
  • LLM routing checkquiz - 6 min