Having read or at least skimmed most of those books, I think the best intro is 'CUDA Programming: A Developer's Guide to Parallel Computing with GPUs'

Massively Parallel Processors: A Hands-on Approach is not really good in my opinion, many small mistakes and confusing sentences (even when you know cuda).

CUDA by Example: An Introduction to General-Purpose GPU Programming is too simple and abstract too much the architecture.

Next year I'm planning to start writing a cuda book that starts by engineering the hardware, and goes up to the optimization part on that harware (which is basically a nvidia card) including all the main algorithms (except for graphs).

I'm already teaching the course in this way at uni, and it is quite successful among students.

Regarding the section on Python and high-level CUDA, anyone interested should maybe first take a peek at Warp, which I’m guessing is too new to have a book yet. Warp lets you write CUDA kernels directly in Python, and it’s a breeze to get started. https://github.com/nvidia/warp

Does anyone know of any good resources for the newer paradigms like cuTile?

Increasingly (for instance ADSP podcast [1]) those in nvidia's inner circle are advocating against writing your own CUDA kernels. (Unless that's your full time job at nvidia, that is).

[1] https://adspthepodcast.com/2024/08/30/Episode-197.html

"AI Systems Performance Engineering" might deserve a mention, even though it's not strictly CUDA.

I liked going through https://www.olcf.ornl.gov/cuda-training-series/ for an intro and some fundamentals.

First one I clicked on is 404: Programming Massively Parallel Processors: A Hands-on Approach (3rd Edition) https://www.cambridge.org/core/books/programming-in-parallel...

Any good MOOCs on Parallel programming/NVIDIA?

In an age when your company mandates you to raise your productivity right now with hundreds of percentage points using LLMs, how do you find an excuse to sit down and read a book?

Having read or at least skimmed most of those books, I think the best intro is 'CUDA Programming: A Developer's Guide to Parallel Computing with GPUs'

Massively Parallel Processors: A Hands-on Approach is not really good in my opinion, many small mistakes and confusing sentences (even when you know cuda).

CUDA by Example: An Introduction to General-Purpose GPU Programming is too simple and abstract too much the architecture.

I'm already teaching the course in this way at uni, and it is quite successful among students.

I really wish there were better options to PMPP... It's by far the most up-to-date book, but I totally agree the writing is sort of bad and some of the code examples are straight up incorrect.

So tl;dr, you have at least one person who would pay for a better book :-)

the first book was published in 2012,is it too outdated?

I really wish there were better options to PMPP... It's by far the most up-to-date book, but I totally agree the writing is sort of bad and some of the code examples are straight up incorrect.

So tl;dr, you have at least one person who would pay for a better book :-)

"AI Systems Performance Engineering" might deserve a mention, even though it's not strictly CUDA.

Increasingly (for instance ADSP podcast [1]) those in nvidia's inner circle are advocating against writing your own CUDA kernels. (Unless that's your full time job at nvidia, that is).

[1] https://adspthepodcast.com/2024/08/30/Episode-197.html

can very much agree about not writing stuff like reductions yourself, unless you have good reason to. but this sort of feels like another "implement everything with <nvidia stuff> and you'll have a great time!! (but also coincidentally get locked in even more to Nvidia hardware)"

That advice seems like nonsense. It's like saying avoid C because you can use Python, or avoid writing a graphics engine because you can license Unreal.

That would be cool but nvidia released blackwell and still have not released unbroken kernels for sm120. Sm120 is not the data center gpu, so it doesn't get its love. So we can't depend on nvidia to do the right thing is my point unfortunately

It’s not about whether you work at Nvidia. Avoid writing CUDA kernels if there are higher level libraries that do what you need. Do write CUDA kernels if you want to learn how, or if you need the low level control, or to micro-optimize. Being able to fuse kernels to avoid memory traffic or get better specialization is also a reason to reach for raw CUDA. Just consider what’s the right tool for the job…

I liked going through https://www.olcf.ornl.gov/cuda-training-series/ for an intro and some fundamentals.

Going through books after this one was a breeze

First one I clicked on is 404: Programming Massively Parallel Processors: A Hands-on Approach (3rd Edition) https://www.cambridge.org/core/books/programming-in-parallel...

the newest is 4th ed i think

In an age when your company mandates you to raise your productivity right now with hundreds of percentage points using LLMs, how do you find an excuse to sit down and read a book?

It feels like a dirty secret, doesn't it?

Don't you read while your agents are doing all the work for you? /s

That advice seems like nonsense. It's like saying avoid C because you can use Python, or avoid writing a graphics engine because you can license Unreal.

Going through books after this one was a breeze

the newest is 4th ed i think

the first book was published in 2012,is it too outdated?

Not really, Hardware didn't really change that much, of course you'll not find Tensor or raytracing cores, but you will have a very solid grasp of gpu programming and the cuda language (that didn't change that much either), and then you can easily learn those more modern things with blog posts or even, at worst, chatgpt.

It feels like a dirty secret, doesn't it?

Yeah, corps don't want you to know how to code, they want you to be a prompter...

Don't you read while your agents are doing all the work for you? /s

Or make your agents do the reading for you!

Yeah, corps don't want you to know how to code, they want you to be a prompter...

Or make your agents do the reading for you!

Any good MOOCs on Parallel programming/NVIDIA?

Does anyone know of any good resources for the newer paradigms like cuTile?

Awesome CUDA Books

A curated list of every major book on CUDA programming — beginner to advanced, C++/Python, architecture, optimization, and the latest 2024–2026 releases.
Focused on practical, high-quality resources for NVIDIA GPU parallel computing.

Contributions welcome! See Contributing.

Beginner / Getting Started
Core Architecture & Parallel Programming
Practical & Hands-on Guides
Advanced / Optimization / Reference
Python & High-Level CUDA
Modern & Recent Releases (2022–2026)
Contributing
Related Awesome Lists

Beginner / Getting Started

CUDA by Example: An Introduction to General-Purpose GPU Programming
Jason Sanders & Edward Kandrot (2010, Addison-Wesley)
The timeless classic. Short, example-driven, perfect first book.
Learn CUDA Programming
Jaegeun Han & Bharatkumar Sharma (2019, Packt)
Modern beginner-to-intermediate with CUDA 10+ examples and GitHub repo.
CUDA for Engineers: An Introduction to High-Performance Parallel Computing
Mete Yurtoglu & Duane Storti (2016, Addison-Wesley)
Engineer-focused, hands-on projects for scientists and non-CS folks.

Core Architecture & Parallel Programming

Programming Massively Parallel Processors: A Hands-on Approach (3rd Edition)
David B. Kirk & Wen-mei W. Hwu (2022)
The definitive GPU architecture bible. Used in universities worldwide.

Practical & Hands-on Guides

Programming in Parallel with CUDA: A Practical Guide
Richard Ansorge (2022, Cambridge University Press)
Real-world scientific examples (stencils, Monte Carlo, imaging). Excellent modern C++ coverage.
Professional CUDA C Programming
John Cheng, Max Grossman & Ty McKercher (2014, Wrox)
Production-level: multi-GPU, streams, libraries, and performance pitfalls.
GPU Parallel Program Development Using CUDA
Tolga Soyata (2018, Chapman & Hall/CRC)
Strong on libraries (cuBLAS, cuFFT, Thrust, NPP) and OpenCL comparison.
CUDA for Deep Learning Elliot Arledge (2025, Manning) From first kernels to Flash Attention — teaches hands-on CUDA optimization for deep learning with Nsight Compute profiling.

Advanced / Optimization / Reference

The CUDA Handbook: A Comprehensive Guide to GPU Programming
Nicholas Wilt (2013)
The deep-dive reference. Every API detail and low-level trick.
CUDA Programming: A Developer's Guide to Parallel Computing with GPUs
Shane Cook (2013, Morgan Kaufmann)
Parallel algorithms, optimization patterns, and best practices.
CUDA Application Design and Development
Rob Farber (2011, Morgan Kaufmann)
Real research applications and scalable design.

Python & High-Level CUDA

Hands-On GPU Programming with Python and CUDA
Brian Tuomanen (2018, Packt)
Best for Python users — Numba, CuPy, and raw bindings.
GPU Programming with C++ and CUDA (or 9781805128823 variant)
Paulo Motta (2024, Packt)
Modern C++20 + Python interop (pybind11).

Modern & Recent Releases (2022–2026)

Programming in Parallel with CUDA (Ansorge, 2022) — see above
Programming Massively Parallel Processors (3rd Ed.) (Kirk & Hwu, 2022) — see above
GPU Programming with C++ and CUDA (Motta, 2024) — see above
CUDA for Deep Learning (Arledge, 2025, Manning) — see above

Notable 2024–2026 titles (mostly specialized or self-published but frequently appearing in searches):

CUDA C++ Optimization – David Spuler (2024) — kernel performance & memory tuning
CUDA C++ Debugging – Dr. David Spuler (2024) — error checking & Nsight
CUDA Programming from Basics to Advanced – Finbarrs Oketunji (2024, covers CUDA 12.6)
CUDA Mastery – Elbert Gale (2024) — scientific simulations & CUDA-X
CUDA in Action – Leon Chapman (2024) — Tensor Cores & multi-GPU
Mastering CUDA C++ Programming – Brett Neutreon (2024) / Toby Webber (2025) — comprehensive C++ guides
High-Performance Computing with C++26 and CUDA 13 – William M. Crutcher (2026)

Pro tip: CUDA changes fast. Always pair books with the free official CUDA C++ Programming Guide (v13.x, 2026).

Contributing

Contributions are welcome! See contributing.md for the full guide.

Quick version:

Add a new high-quality book? Open a PR with title, authors, year, short description, and link.
Preference for books post-2018 or still relevant classics.
Only include books with substantial code/examples and good reviews.

Related Awesome Lists

Star the repo if this helps you write faster kernels! 🚀

Inspired by GoBooks.

Hacker Times