menu
{ "item_title" : "Inside DeepInfer", "item_author" : [" Yang Guo "], "item_description" : "This book walks you through DeepInfer, the inference engine DeepSeek uses to serve DeepSeek-V4.1-Flash in production. By the end you should be able to explain how a request is scheduled, computed and returned inside the engine, and why the system is designed the way it is. Who this book is forEngineers working on inference systems, model deployment or GPU performance.Students who know Transformers and want to learn how a trained model is served efficiently.Readers who have studied vLLM or SGLang and want to see the trade-offs a large production system makes.Software engineers with only a high-level view of ML, and hardware engineers without a CS background. Chapter 1 builds up every prerequisite term from scratch.Prerequisites: you can read Python and basic C++. No CUDA experience is needed.", "item_img_path" : "https://covers4.booksamillion.com/covers/bam/9/79/816/489/9798164896539_b.jpg", "price_data" : { "retail_price" : "19.99", "online_price" : "19.99", "our_price" : "19.99", "club_price" : "19.99", "savings_pct" : "0", "savings_amt" : "0.00", "club_savings_pct" : "0", "club_savings_amt" : "0.00", "discount_pct" : "10", "store_price" : "" } }
Inside DeepInfer|Yang Guo

Inside DeepInfer : A Guide to DeepSeek's Large-Scale Inference Engine

local_shippingShip to Me
In Stock.
FREE Shipping for Club Members help

Overview

"This book walks you through DeepInfer, the inference engine DeepSeek uses to serve DeepSeek-V4.1-Flash in production. By the end you should be able to explain how a request is scheduled, computed and returned inside the engine, and why the system is designed the way it is.

Who this book is for
Engineers working on inference systems, model deployment or GPU performance.Students who know Transformers and want to learn how a trained model is served efficiently.Readers who have studied vLLM or SGLang and want to see the trade-offs a large production system makes.Software engineers with only a high-level view of ML, and hardware engineers without a CS background. Chapter 1 builds up every prerequisite term from scratch.
Prerequisites: you can read Python and basic C++. No CUDA experience is needed."

This item is Non-Returnable

Details

  • ISBN-13: 9798164896539
  • ISBN-10: 9798164896539
  • Publisher: Independently Published
  • Publish Date: October 2026
  • Dimensions: 9 x 6 x 0.21 inches
  • Shipping Weight: 0.32 pounds
  • Page Count: 102

Related Categories

You May Also Like...

    1

BAM Customer Reviews