Deployment & Scaling · Advanced
Bedrock Cost and Scale: Caching, Batch, Throughput and Cross-Region Inference
Five levers that change generative AI unit economics on AWS by an order of magnitude — and how to attribute the spend when finance asks.
- 45 min lesson
- advanced level
- AWS AI Track
- Narrated presentation
What you’ll learn
- Model cascading as the dominant cost lever
- Implicit vs explicit prompt caching, checkpoint order and token minimums
- Cache-read discounts, cache-write premiums and rate-limit exemption
- Batch inference economics and its incompatibility with caching
- Geographic vs global cross-Region inference; Provisioned Throughput break-even
- Tagging via application inference profiles and reading cache usage fields
Inside this lesson
A narrated, slide-by-slide presentation with 1 prediction exercise, 1 knowledge check, 5 videos. You answer as you go, and the platform checks your work.