Apple M4 GPU (10-Core) vs A18 Pro GPU

We compared two integrated laptop graphics cards: the Apple M4 GPU (10-Core) with 160 pipelines and 1280 shaders against the 4 months newer A18 Pro GPU that utilizes 128 pipelines and 768 shaders. Here you will find complete details about specs, efficiency, performance tests, and more.

Review

General comparison of performance in games, applications, power efficiency, and other metrics
Gaming
Performance in DirectX, OpenCL, and Vulkan games
Workstation
Perf. in 3D modeling, video editing and rendering apps
AI/ML
Capabilities for machine learning and AI-related tasks
Energy Efficiency
Power consumption efficiency in different scenarios
NanoReview Final Score
Overall video card score

Key differences

Key distinctions and advantages of A18 Pro GPU over M4 GPU (10-Core)
Reasons to consider the Apple M4 GPU (10-Core)
  • Performs significantly better (up to 2.3x) in 3DMark Steel Nomad Lite
  • 65% higher maximum theoretical performance (3.8 vs 2.3 TFLOPS)
  • Achieves 21% more points in the GeekBench 6 Compute test (37K vs 31K)
  • Has 67% more shading units (1280 vs 768)

Benchmarks

Graphics cards’ performance in recent benchmarking apps

3D Mark

Multiplatform graphics benchmark suite that directly correlates with performance in modern games
Steel Nomad Lite Score
4103
Solar Bay 16554 7374
Wild Life Extreme 9610 3947
Sources: 3DMark [1], [2]

GeekBench 6 GPU Compute

GPU test for computational tasks (image processing, photography, computer vision, and ML)
GB6 Compute Score
37678
31229
Background Blur 72.2 img/sec 57.1 img/sec
Face Detection 48.3 img/sec 91.4 img/sec
Horizon Detection 1.4 Gpixels/sec 0.92 Gpixels/sec
Edge Detection 1.94 Gpixels/sec 1.21 Gpixels/sec
Gaussian Blur 1.5 Gpixels/sec 1.5 Gpixels/sec
Feature Matching 0.53 Gpixels/sec 0.45 Gpixels/sec
Stereo Matching 123.3 Gpixels/sec 84.2 Gpixels/sec
Particle Physics 4938.4 FPS 3061 FPS
API OpenCL Metal
Sources: Geekbench [3], [4]

Cinebench 2024 GPU

Hardware benchmark using Maxon's Cinema 4D rendering engine
Cinebench 2024 GPU

Blender

Rendering performance test for 3D modeling
Blender GPU
Sources: Blender [9]529 samples

Recent User Tests

The latest benchmark tests that have been submitted by users
Apple M4 GPU (10-Core)
No benchmark results yet
Apple A18 Pro GPU
DateBenchmarkResult
📘 2026-04-24 -> RennsportGeekBench19670

Artificial Intelligence Tests

Performance in machine learning and artificial intelligence tasks

GeekBench 6 ML

Tests throughput of AI operations in single, half, and quantized precision
GB6 ML Single Precision
10346
GB6 ML Half Precision
11883
GB6 ML Quantized
11036
Image Classification (SP) 4874 3264
Image Segmentation (HP) 8168 4550
Image Super Resolution (Q) 11983 6705
Face Detection (HP) 17036 11770
Pose Estimation (Q) 31171 15548
Text Classification (SP) 2964 3678
Machine Translation (HP) 3410 5678
Object Detection (SP) 5112 3360
Depth Estimation (Q) 22700 13468
Style Transfer (SP) 60872 31541
Framework Core ML Core ML
Backend GPU GPU
Sources: Geekbench [9], [10]

Specifications

Technical specifications of Apple M4 GPU (10-Core) and A18 Pro GPU

General

Vendor Apple Apple
Build Integrated Integrated
Released May 7, 2024 September 9, 2024
Case Laptop Laptop
Purpose Professional Generic
Segment Mid-range Mid-range
Architecture Apple M GPU Apple M GPU
GPU Codename Custom Custom
Rival Equivalent - Adreno X1-85 -
Successor - Apple M5 GPU (10-Core) -
Recommended CPU - Apple M4 (10-Core) or above - Apple A18 Pro (MacBook) or above
Used in CPUs - Apple M4 (10-Core) - Apple A18 Pro (MacBook)
Laptop GPU ranking (71st and 81st place)

Graphics Processing Unit

Base Clock 500 MHz -
Boost Clock 1470 MHz 1490 MHz
Shading Units 1280 768
Texture Mapping Units (TMUs) 80 40
Render Output Units (ROPs) 40 -
Compute Units (Pipelines) 160 128
Instructions Per Cycle 2 IPC 2 IPC

Raw Performance

Pixel Fill Rate 59 GPixel/s -
Texture Fill Rate 118 GTexel/s 60 GTexel/s
FLOPS (FP32)
3.8 TFLOPS
2.3 TFLOPS

Physical

Interface Custom Custom
TGP 18 W -
Manufacturing TSMC TSMC
Fabrication Process 3 nm 3 nm
Transistor Count 28 billion -
Max. Temperature 100°C -

Memory

Memory Type System Shared System Shared
Memory Clock 7500 MHz 7500 MHz
Effective Memory Speed 15000 Mbps -
Bus 128-bit 128-bit
ECC No No
Memory Bandwidth

API

Ray Tracing Yes Yes
DLSS No No

Cast your vote

Choose between two graphics cards
3 (75%)
1 (25%)
Total votes: 4

User opinions

You can share your opinion or ask a question in the comments below
🌐 Register your profile and become part of NanoReview community!