CUDA for LLMs
Sorry, there was a problem loading this page. Try again.
{"desktop_buybox_group_1":[{"displayPrice":"$66.99","priceAmount":66.99,"currencySymbol":"$","integerValue":"66","decimalSeparator":".","fractionalValue":"99","symbolPosition":"left","hasSpace":false,"showFractionalPartIfEmpty":true,"offerListingId":"v3req8EzS2rEEkmd5JIo4NQtLamRuCMWe215RViK1X9HPM2SZdVe1TBVLiat6jTpthZmeQv6iUqvVYG6wHVgHpNFvNM6n0PizbYJXOdjxvfuk7LZDn%2BOzwwG09k6hzOaRTeNrm4skVYd512jbfX0hw%3D%3D","locale":"en-US","buyingOptionType":"NEW","aapiBuyingOptionIndex":0}]}
Purchase options and add-ons
Get the eBook free when you register your print book at Manning.
CUDA (Compute Unified Device Architecture) provides a powerful parallel programming model AI engineers can use to tap the massive processing power of NVIDIA GPUs. CUDA delivers direct control, debugging power, and acceleration at the GPU level that canât be matched by other types of optimizations.
This book shows you how to work within the CUDA ecosystem, from your first kernel to implementing advanced LLM features like Flash Attention. Youâll learn to profile with Nsight Compute, identify bottlenecks, and understand why each optimization works. By solving problems at multiple levels of abstraction, youâll develop a deep understanding of CUDA, along with a practical mastery of kernel-building skills. Written for the latest NVIDIA hardware, the book builds a deep understanding of CUDA fundamentals that will stay relevant as chips upgrade and evolve.
What's inside
âĒ 56 kernels to utilize in your models
âĒ PyTorch C++ extension pipeline for integrating custom kernels
âĒ Exploit advanced NVIDIA GPU features (Ampere, Hopper, Blackwell)
âĒ Build backpropagation from scratch, ending with a single-file MNIST MLP
About the reader
For software and AI engineers comfortable with C/C++. No prior CUDA experience required.
About the author
Elliot Arledge created the 12-hour CUDA course and the 6-hour LLM from Scratch course for FreeCodeCamp, and consults on deep learning performance.
CUDA (Compute Unified Device Architecture) provides a powerful parallel programming model AI engineers can use to tap the massive processing power of NVIDIA GPUs. CUDA delivers direct control, debugging power, and acceleration at the GPU level that canât be matched by other types of optimizations.
This book shows you how to work within the CUDA ecosystem, from your first kernel to implementing advanced LLM features like Flash Attention. Youâll learn to profile with Nsight Compute, identify bottlenecks, and understand why each optimization works. By solving problems at multiple levels of abstraction, youâll develop a deep understanding of CUDA, along with a practical mastery of kernel-building skills. Written for the latest NVIDIA hardware, the book builds a deep understanding of CUDA fundamentals that will stay relevant as chips upgrade and evolve.
What's inside
âĒ 56 kernels to utilize in your models
âĒ PyTorch C++ extension pipeline for integrating custom kernels
âĒ Exploit advanced NVIDIA GPU features (Ampere, Hopper, Blackwell)
âĒ Build backpropagation from scratch, ending with a single-file MNIST MLP
About the reader
For software and AI engineers comfortable with C/C++. No prior CUDA experience required.
About the author
Elliot Arledge created the 12-hour CUDA course and the 6-hour LLM from Scratch course for FreeCodeCamp, and consults on deep learning performance.
- ISBN-101633434893
- ISBN-13978-1633434899
- PublisherManning
- Publication dateDecember 1, 2026
- LanguageEnglish
- Print length425 pages
Customers who viewed this item also viewed
Page 1 of 1 Start over
- Programming Massively Parallel Processors: A Hands-on ApproachPaperbackFREE Shipping by AmazonGet it as soon as Wednesday, Nov 4
- Programming Massively Parallel Processors: A Hands-on ApproachPaperbackFREE Shipping by AmazonGet it as soon as Wednesday, Nov 4Only 18 left in stock - order soon.
- GPU Programming with C++ and CUDA: Uncover effective techniques for writing efficient GPU-parallel C++ applicationsPaperbackFREE Shipping by AmazonGet it as soon as Wednesday, Nov 4
- AI Agents in Action, Second Edition: Intelligent workflows with LLMs, MCP, A2A, and morePaperbackFREE Shipping by AmazonGet it as soon as Wednesday, Nov 4
- GPU-Accelerated Computing with Python 3 and CUDA: From low-level kernels to real-world applications in scientific computing and machine learningPaperbackFREE Shipping by AmazonGet it as soon as Wednesday, Nov 4
Customers also bought or read
Page 1 of 1Start over
- Build a Reasoning Model (From Scratch)#1 Best SellerProgramming Algorithms
Paperback$47.99$47.99FREE delivery Wed, Nov 4 - Build a Large Language Model (From Scratch)#1 Best SellerComputer Neural Networks
Paperback$47.44$47.44FREE delivery Wed, Nov 4 - Domain-Specific Small Language Models: Efficient AI for local deployment
Paperback$47.23$47.23$3.99 delivery Nov 13 - 18 - Reinforcement Learning from Human Feedback: LLM alignment and post-training
Just releasedPaperback$59.99$59.99FREE delivery Wed, Nov 4 - Build an AI Agent (From Scratch): Agents that reason, plan, and act autonomously
Just releasedPaperback$57.04$57.04FREE delivery Wed, Nov 4 - Deep Learning with PyTorch, Second Edition: Training and applying deep learning and generative AI models
Paperback$51.68$51.68FREE delivery Wed, Nov 4 - AI Engineering: Building Applications with Foundation Models#1 Best SellerEnterprise Applications
Paperback$46.00$46.00FREE delivery Wed, Nov 4 - Design Multi-Agent AI Systems Using MCP and A2A: Engineer your own Python-based agentic AI framework with tool use, memory, and multi-agent workflows
Paperback$38.49$38.49FREE delivery Wed, Nov 4 - Agentic Spec-Driven Development: A Practical Method for Using AI to Build Complete Specifications for Software, Products, and Knowledge Work
Paperback$39.95$39.95FREE delivery Wed, Nov 4 - Distributed AI Systems: A practical guide to building scalable training, inference, and serving systems for production AI
Paperback$31.99$31.99Delivery Wed, Nov 4 - AI Agents: The Definitive Guide: Design, Deployment, and Evaluation for Production
Just releasedPaperback$53.35$53.35FREE delivery Wed, Nov 4 - Agentic Coding with Claude Code: The everyday developer's guide to agentic coding with Claude Code
Paperback$32.78$32.78Delivery Wed, Nov 4 - System Design for the LLM Era: Patterns and principles for production-grade AI architecture
Paperback$33.24$33.24Delivery Wed, Nov 4 - Evals for AI Engineers: Systematically Measuring and Improving AI Applications
PreorderPaperback$70.10$70.10FREE delivery Nov 24 - 30 - Building Claude Skills: Turn repeatable work into reusable Claude Skills#1 Best SellerSoftware Reuse
Paperback$23.74$23.74Delivery Wed, Nov 4 - Hands-On LLM Serving and Optimization: Hosting LLMs at Scale
Paperback$60.29$60.29FREE delivery Wed, Nov 4 - Reinforcement Learning, second edition: An Introduction (Adaptive Computation and Machine Learning series)
Hardcover$66.66$66.66FREE delivery Wed, Nov 4 - LLMs in Production: From language models to successful products
Paperback$48.13$48.13FREE delivery Wed, Nov 4 - Build with Claude Code: From zero to deployed apps with AI-assisted development#1 New ReleaseEnterprise Applications
Paperback$34.99$34.99Delivery Wed, Nov 4 - Machine Learning with PyTorch and Scikit-Learn: Develop machine learning and deep learning models with Python#1 Best SellerSpeech & Audio Processing
Paperback$34.23$34.23Delivery Wed, Nov 4 - Hands-On Machine Learning with Scikit-Learn and PyTorch: Concepts, Tools, and Techniques to Build Intelligent Systems
Paperback$66.23$66.23FREE delivery Wed, Nov 4 - Deep Reinforcement Learning Hands-On: A practical and easy-to-follow guide to RL from Q-learning and DQNs to PPO and RLHF
Paperback$29.65$29.65Delivery Wed, Nov 4 - Building Reliable AI Systems: Applications and agents you can trust
Just releasedPaperback$45.70$45.70FREE delivery Wed, Nov 4
Loading...
Editorial Reviews
About the Author
Elliot Arledge created the 12-hour CUDA course and the 6-hour LLM from Scratch course for FreeCodeCamp, and consults on deep learning performance.
Product details
- Publisher : Manning
- Publication date : December 1, 2026
- Language : English
- Print length : 425 pages
- ISBN-10 : 1633434893
- ISBN-13 : 978-1633434899
- Item Weight : 15.8 ounces
- Best Sellers Rank: #1,361,517 in Books (See Top 100 in Books)
- #306 in C++ Programming Language
- #2,294 in Computer Programming Languages
Filtering customer reviews
Loading customer reviews
Sign in to see customer reviews.















