What CUDA Actually Is
NVIDIA's platform for running C++ on the GPU.
What CUDA Actually Is is a free CUDA Academy lesson on CoddyKit — lesson 3 of 4. You can read the complete lesson below for free — then practise it hands-on in the browser with a built-in code editor and a 24/7 AI tutor. It is part of the CUDA Academy learning path, one of 4 lessons in the course, and your progress syncs across the web and the CoddyKit app.
Putting a Name to the Magic
You have seen why GPUs are fast. CUDA is the tool that lets you actually write code for them in C++. Time to meet it properly. 🚀
CUDA Is a Platform
CUDA is NVIDIA's platform for general-purpose GPU computing. It lets you run your own programs on the GPU, not just graphics.
An Extension of C++
CUDA C++ is just C++ with a few extra keywords and functions. If you know C++, most of CUDA already feels familiar from day one.
Kernels: Code That Runs on the GPU
A function you mark to run on the GPU is called a kernel. Thousands of threads will each run that one function in parallel.
__global__ void add(int *out, int *a, int *b) {
// runs on the GPU
}The __global__ Keyword
That __global__ tag marks a function as a kernel: callable from the CPU but executed on the GPU. It is your gateway onto the device.
Host and Device
In CUDA talk, the CPU side is the host and the GPU side is the device. You will hear these two words constantly, so lock them in now.
The Runtime API
CUDA gives you a friendly runtime API: functions like cudaMalloc and cudaMemcpy that manage GPU memory and move data around for you.
nvcc: The CUDA Compiler
You build CUDA code with nvcc. It splits your file, sends host code to your normal compiler and device code to the GPU compiler.
It Is NVIDIA-Only
CUDA runs only on NVIDIA GPUs. That tight coupling is why it is so well tuned, but it also means it is vendor specific.
A Rich Library Ecosystem
CUDA ships tuned libraries like cuBLAS for matrices and cuDNN for deep learning, so you rarely start a hard problem from scratch.
Why CUDA Took Over
Mature tools, fast libraries, and deep ecosystem support made CUDA the default way to do GPU computing in science and AI today.
Quick Check
Let us confirm what CUDA really is.
Recap: What CUDA Is
CUDA is NVIDIA's platform that extends C++ with kernels, a runtime API, and the nvcc compiler so your code can run on the GPU. 👍
Frequently asked questions
Is the “What CUDA Actually Is” lesson free?
Yes — the full text of “What CUDA Actually Is” is free to read here on the web, and the CUDA Academy course includes 4 lessons in total. To practise it interactively (a built-in code editor and a 24/7 AI tutor) and unlock the rest of the CUDA Academy course, upgrade to CoddyKit PRO.
What will I learn in “What CUDA Actually Is”?
NVIDIA's platform for running C++ on the GPU. You practise CUDA Academy with hands-on code you run directly in the browser, and a 24/7 AI tutor answers your questions as you work through the lesson.
Do I need any experience to start CUDA Academy?
No prior experience is required. CUDA Academy on CoddyKit is structured for beginners through advanced learners; this is — lesson 3 of 4, so you can start here or from the beginning and move at your own pace.
How long does the “What CUDA Actually Is” lesson take?
Most CoddyKit lessons take about 5–10 minutes. Each one is bite-sized and interactive, so you make steady progress and pick up exactly where you left off across the web and the app.
Can I write and run code in this CUDA Academy lesson?
Yes. Every CUDA Academy lesson includes a built-in code editor, so you write and run real code right in your browser and get instant AI feedback — no local setup required.