Apple
GPUIntegrated (APU)Released Sep 22, 2026

Apple M6 GPU (12-core)

1,536 cores

GPU block of the Apple M6, introduced in the Mac mini. Every core carries a Neural Accelerator, which Apple credits with nearly 30% more peak GPU compute for AI than M5 and more than 8x M1. The generation also brings an updated shader core architecture and Dynamic Caching, hardware-accelerated ray tracing, and 50% higher geometry rates for complex graphics.

General

Manufacturer
Apple
Model
M6 GPU (12-core)
Segment
Integrated (APU)
GPU chip
Apple M6
Architecture
Apple GPU
Release date
Sep 22, 2026
Price
Search on Amazon

Memory

Memory size
—
Memory type
LPDDR5X (unified)
Bus width
—
Bandwidth
170 GB/s
Memory clock
—

Clocks & silicon

GPU clock
—
Boost clock
—
Shader units
1,536 ALUs (12 cores)
Tensor / AI cores
—
RT cores
—
Transistors
—
Technology (nm)
2 nm
TSMC N2
TDP / TBP
—

Interface & APIs

Bus interface
Integrated (on-die)
SLI / multi-GPU
None
DirectX
N/A (macOS / iPadOS)
OpenGL
4.1 (deprecated on macOS)
Vulkan
via MoltenVK
Other APIs
Metal 4, hardware ray tracing, updated Dynamic Caching, Neural Accelerator per GPU core

AI support

Precisions
FP8FP16BF16INT8INT4TF32FP64
AI accelerator
Dual 16-core Neural Engine (SoC) + 12 GPU Neural Accelerators
Peak AI TOPS
—
Notes
GPU supports FP16/INT8/BF16 via Metal. A Neural Accelerator sits in every GPU core. The Dual 16-core Neural Engine is a separate SoC block. Apple does not publish GPU clocks or TOPS for M6.

Related Apple GPUs

All Apple GPUs →

Integrated (APU) gpus from Apple announced around the same time as the Apple M6 GPU (12-core). Add any of them to your compare list to see the specs side by side.