Context Caching for Faster and Cheaper LLM Inference

Explore context caching techniques for faster and cheaper inference in large language models (LLMs) in this 35-minute video tutorial. Learn how context caching works, its two main types, and implementation strategies for Claude, Google Gemini, and SGLang. Discover the cost-saving potential of this advanced inference technique and gain practical insights into improving LLM performance. Access comprehensive resources, including code repositories, slides, and timestamps, to enhance your understanding of this cutting-edge topic in AI development.

Introduction to context caching for LLMs
Video Overview
How does context caching work?
Two types of caching
Context caching with Claude and Google Gemini
Context caching with Claude
Context caching with Gemini Flash or Gemini Pro
Context caching with SGLang works also with vLLM
Cost Comparison
Video Resources