# Layer 2 · Compute Hardware

URL: https://justuju.org/layer/compute-hardware/

The chips that do the maths: GPUs and other accelerators, their fast memory, and the links between them.

Gives the layer above: Raw arithmetic throughput and memory, exposed to software through platforms such as CUDA or ROCm.
Needs from the layer below: Power, cooling and fabricated silicon from Physical & Energy.

## Concepts

- [GPUs & AI accelerators](https://justuju.org/concept/gpu-accelerators/): Chips built to do enormous numbers of simple multiplications in parallel, the core operation of neural networks.
- [Number formats (FP16, BF16, FP8, FP4)](https://justuju.org/concept/number-formats/): How many bits each number gets inside the chip. Fewer bits mean more maths per second and more numbers per trip from memory, at some cost in precision.
- [Memory & bandwidth (HBM)](https://justuju.org/concept/memory-bandwidth/): How fast data reaches the chip often limits AI speed more than the chip's raw arithmetic.
- [Chip-to-chip interconnect](https://justuju.org/concept/chip-interconnect/): The links that let several accelerators share work. Fast dedicated links let a group of chips act like one big chip; slower general-purpose links make them wait.
- [Local AI hardware](https://justuju.org/concept/local-ai-hardware/): Running models on your own machine: consumer graphics cards, multi-GPU home rigs, or desktop machines with large unified memory. Memory size decides what fits; memory speed decides how fast it runs.

---
Justuju · https://justuju.org/ · Content CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote with a link to the page.