Tensor G5 vs A18 Pro

We compared the 8-core Google Tensor G5 (PowerVR DXT-48-1536) with the older 6-core Apple A18 Pro (Apple A18 Pro GPU) SoC. Here you will find the pros and cons of each chip, technical specs, and comprehensive tests in benchmarks, like AnTuTu and Geekbench.

Review

General comparison of performance, power consumption, and other indicators
CPU Performance
Single and multi-core processor tests
Gaming Performance
GPU performance in games and OpenCL/Vulkan
Battery life
Efficiency of battery consumption
NanoReview Score
Overall chip score

Key Differences

Main differences and advantages of each chip
Pros of Google Tensor G5
  • Has 2 more cores
  • Supports 14% higher memory bandwidth (68.2 against 60 GB/s)
Pros of Apple A18 Pro
  • Higher GPU frequency (~35%)
  • Performs 35% better in floating-point computations
  • 7% higher CPU clock speed (4050 vs 3780 MHz)
  • Shows slightly better (up to 15%) AnTuTu 11 score – 1782K vs 1549K

Benchmarks

Performance tests in popular benchmarks
SoC:
vs

AnTuTu 11

The AnTuTu Benchmark measures CPU, GPU, RAM, and I/O performance in different scenarios
1549798
A18 Pro +15%
1782199
CPU 734567 521852
GPU 183228 562674
Memory 222379 187942
UX 409624 509731
Total score 1549798 1782199
Samples [16 tests] [15 tests]
AnTuTu results from iOS and Android are not directly comparable! They won't affect NanoReview scores.

GeekBench 6

The GeekBench test shows raw single-threaded and multithreaded CPU performance
Single-Core Score
2305
A18 Pro +55%
3570
Multi-Core Score
6054
A18 Pro +47%
8923
Asset compression 281.8 MB/sec 299.9 MB/sec
HTML 5 Browser 162.5 pages/sec 191.2 pages/sec
PDF Renderer 207.4 Mpixels/sec 196.8 Mpixels/sec
Image detection 149.5 images/sec 319.6 images/sec
HDR 151.4 Mpixels/sec 289.5 Mpixels/sec
Background blur 24.4 images/sec 37.6 images/sec
Photo processing 46.5 images/sec 91 images/sec
Ray tracing 6.79 Mpixels/sec 9.03 Mpixels/sec
Samples 8 tests 14 tests
Compute Score (GPU)
4116
A18 Pro +709%
33278
Background blur 4.77 images/sec 62.5 images/sec
Face detection 2.41 images/sec 78.2 images/sec
Horizon detection 0.17 Gpixels/sec 0.98 Gpixels/sec
Edge detection 0.27 Gpixels/sec 1.3 Gpixels/sec
Gaussian blur 0.2 Gpixels/sec 1.5 Gpixels/sec
Feature matching 0.05 Gpixels/sec 0.49 Gpixels/sec
Stereo matching 12.6 Gpixels/sec 100 Gpixels/sec
Particle physics 634.9 FPS 3659.7 FPS
API Vulkan Metal
Samples 7 tests 7 tests
Compute results use different APIs (Metal vs Vulkan/OpenCL) and are not directly comparable!

3DMark

A cross-platform benchmark that assesses graphics performance in Vulkan (Metal)
3DMark Steel Nomad Light
1012
A18 Pro +71%
1732
Score 1012 1732
Stability 96% 71%
Graphics test 7 FPS 13 FPS
3DMark Solar Bay
n/a
7585
Score - 7585
Graphics test - 29 FPS
Section 1 - 35 FPS
Section 2 - 30 FPS
Section 3 - 21 FPS

PCMark 3.0

Benchmark for evaluating overall system performance in everyday workloads
PCMark 3.0
15332
n/a
Web score 17237 -
Video editing score 7203 -
Data manipulation score 14036 -
Photo editing score 24584 -
Writing score 20213 -
Phones tested 4 models -

Recent User Tests

The latest benchmark tests that have been submitted by users
Google Tensor G5
DateBenchmarkResult
πŸ“˜ 2026-10-28 -> PixeluserAnTuTu 121596320
πŸ“˜ 2026-09-25 (Dolbaeb)AnTuTu 121549159
πŸ“˜ 2026-08-22 -> PixeluserAnTuTu 111536042
πŸ“˜ 2026-08-02 -> bangAnTuTu 111719659
πŸ“˜ 2026-08-02 -> bangCPU Throttle74%
Apple A18 Pro
DateBenchmarkResult
πŸ“˜ 2026-10-31 (Rp216)AnTuTu 112084429
πŸ“˜ 2026-10-20 (Idkwhattoputhere)AnTuTu 111880916
πŸ“˜ 2026-09-27 (FontFrontPromptORG)Geekbench 7 CPU3087 / 8369
πŸ“˜ 2026-09-23 -> RappzzzzAnTuTu 112043169
πŸ“˜ 2026-09-20 (thuanday)AnTuTu 112043169

Specifications

Full list of technical specifications of Tensor G5 and A18 Pro

CPU

Architecture 1x 3.78 GHz – Cortex-X4
5x 3.05 GHz – Cortex-A725
2x 2.25 GHz – Cortex-A520
2x 4.05 GHz – Everest
4x 2.42 GHz – Sawtooth
Cores 8 6
Frequency 3780 MHz 4050 MHz
Instruction set ARMv9.2-A ARMv9.2-A
L1 cache - 192 KB
L2 cache - 16 MB
L3 cache 8 MB -
Process 3 nanometers 3 nanometers
TDP (Sustained Power Limit) - 10 W
Manufacturing TSMC TSMC

Graphics

GPU name PowerVR DXT-48-1536 Apple A18 Pro GPU
Architecture - Apple Bionic GPU
GPU frequency 1100 MHz 1490 MHz
Pipelines 6 5
Shading units 128 128
Total shaders 768 640
FLOPS 1689.6 Gigaflops 2289 Gigaflops
Vulkan version 1.4 -
OpenCL version 3.0 -

AI Accelerator

Neural processor (NPU) Yes Neural Engine
NPU theoretical performance - 35 TOPS

Memory

Memory type LPDDR5X LPDDR5X
Memory frequency 4266 MHz 3750 MHz
Bus 4x 16 Bit 4x 16 Bit
Max bandwidth 68.2 Gb/s 60 Gb/s
Max size 16 GB 8 GB

Multimedia (ISP)

Storage type UFS 3.1, UFS 4.0 NVMe
Max display resolution 3840 x 2400 -
Max camera resolution 1x 200MP 1x 48MP
Video capture 8K at 30FPS, 4K at 120FPS 4K at 120FPS
Video playback 8K at 30FPS, 4K at 120FPS 4K at 120FPS
Video codecs - H.264
- H.265
- AV1
- VP9
- H.264
- H.265
- AV1
- VP8
- VP9
- Motion JPEG
Audio codecs - AAC
- AIFF
- CAF
- MP3
- MP4
- WAV
- AAC
- AAX
- AAX+
- AC-3
- AIFF
- CAF
- E-AC-3
- MP3
- MP4
- WAV

Connectivity

Modem Exynos 5400c Snapdragon X71 5G
4G support - LTE Cat. 24
5G support Yes Yes
Download speed - Up to 10000 Mbps
Upload speed - Up to 3500 Mbps
Wi-Fi 7 7
Bluetooth 6.0 5.3
Navigation GPS, GLONASS, Beidou, Galileo, QZSS GPS, GLONASS, Beidou, Galileo, QZSS, NAVIC

Info

Announced August 2025 September 2024
Class Flagship Flagship
Model number GS501 APL1V07
Successor - Google Tensor G6 - Apple A20 Pro

Cast your vote

So, which SoC would you choose?
14 (16.3%)
72 (83.7%)
Total votes: 86

Comments

Please give your opinion on the comparison of A18 Pro and Tensor G5, or ask any questions
🌐 Register your profile and become part of NanoReview community!