Head to head
Compare Graphics Cards
Pick up to 4. Winners are highlighted in green ▲.
1 selected
Key specs
Memory Type
System Shared
Boost Clock
1296 MHz
TGP
120 W
Architecture
Apple M GPU
Benchmarks
Rating
40
Gaming
31
Render
50
Energy
59
TFLOPS (FP32)
21.2
Geekbench OpenCL
106,397
Steel Nomad Lite Score
10,731
GB6 Compute Score
106,397
Cinebench 2024 GPU
3331
Blender GPU
1646.97
GB6 ML Single Precision
13,120
GB6 ML Half Precision
14,103
GB6 ML Quantized
13,229
FLOPS (FP32)
21.2 TFLOPS
Memory Bandwidth
819.2 GB/s
3D Mark
Solar Bay
36,712
Wild Life Extreme
29,343
GeekBench 6 OpenCL
Background Blur
157.3 img/sec
Face Detection
108.7 img/sec
Horizon Detection
5.07 Gpixels/sec
Edge Detection
10.7 Gpixels/sec
Gaussian Blur
6.59 Gpixels/sec
Feature Matching
0.73 Gpixels/sec
Stereo Matching
293.4 Gpixels/sec
Particle Physics
14154.4 FPS
API
OpenCL
GeekBench 6 ML
Image Classification (SP)
4834
Image Segmentation (HP)
14,458
Image Super Resolution (Q)
13,757
Face Detection (HP)
23,618
Pose Estimation (Q)
80,481
Text Classification (SP)
2265
Machine Translation (HP)
1798
Object Detection (SP)
4975
Depth Estimation (Q)
25,755
Style Transfer (SP)
135,609
Framework
Core ML
Backend
GPU
General
Vendor
Apple
Build
Integrated
Released
March 8, 2022
Case
Laptop
Purpose
Professional
Segment
High-end
Architecture
Apple M GPU
GPU Codename
Custom
Nvidia Equivalent
- GeForce RTX 4070 Laptop
Recommended CPU
- Apple M1 Ultra or above
Used in CPUs
- Apple M1 Ultra
Graphics Processing Unit
Base Clock
450 MHz
Boost Clock
1296 MHz
Shading Units
8192
Texture Mapping Units (TMUs)
512
Render Output Units (ROPs)
256
Compute Units (Pipelines)
1024
Instructions Per Cycle
2 IPC
Raw Performance
Pixel Fill Rate
332 GPixel/s
Texture Fill Rate
664 GTexel/s
Physical
Interface
Custom
TGP
120 W
Manufacturing
TSMC
Fabrication Process
5 nm
Transistor Count
92.8 billion
Max. Temperature
94°C
Memory
Memory Type
System Shared
Memory Clock
6400 MHz
Effective Memory Speed
12800 Mbps
Bus
1024-bit
ECC
No
API
Ray Tracing
No
DLSS
No