Skip to content
BENIBIX Learn

Unified Memory

Infrastructure🔧 HardwareMemory

Also known as Unified Memory Architecture, Shared CPU and GPU Memory

One pool of memory that the CPU and the GPU both use. Nothing is copied.

Think of it like

Think of it like one big table that two cooks share. No food is carried from one to the other.

Example

Emma's Mac has 64 GB of unified memory. She loads an AI model that needs 40 GB, which her old 24 GB graphics card could not hold.

Why it matters

Copying data to the GPU takes time and uses the space twice. A shared pool saves both.

Definition

Unified memory is one pool of memory that the CPU and the GPU both use. In most computers, the GPU has its own video memory. Data is copied across to it. With unified memory there is no copy, because both chips read the same bytes. The memory sits right beside the chip, so it usually cannot be added to later. A big pool lets a home computer load AI models too large for most graphics cards.
On the left, data is copied to the GPU's own memory. On the right, both chips read one shared pool.

Common mistake

Counting all of it as room for the GPU. The system and every open app use the same pool, so part of it is always taken.

Key takeaways

  1. 1
    One pool. The CPU and the GPU use the same memory.
  2. 2
    No copies. Data does not have to cross to the GPU.
  3. 3
    Fixed when bought. It cannot be added to later.

Compare

Learn it properly