What is GPU?
The chip designed for graphics that turned out to be perfect for training AI models.
// Explanation
The GPU was created for video games — processing millions of pixels in parallel. It turns out that training AI models requires exactly the same kind of massive parallel math. That's why NVIDIA dominates the AI market: their GPUs (H100, A100) became the 'gold' of the AI era. Without enough GPUs, you can't train large models.
// Example
Training GPT-4 required thousands of NVIDIA GPUs running for weeks. Running a small model like Llama 7B locally needs a GPU with ~8GB VRAM.