Home > Smartphone processors comparison > Tensor G3 vs A17 Pro – what's better?

Tensor G3 vs A17 Pro

Tensor G3
VS
A17 Pro
Tensor G3
A17 Pro

We compared the 9-core Google Tensor G3 (Mali-G715 MP7) with the older 6-core Apple A17 Pro (Apple A17 GPU) SoC. Here you will find the pros and cons of each chip, technical specs, and comprehensive tests in benchmarks, like AnTuTu and Geekbench.

  1. Review
  2. Differences
  3. Benchmarks
  4. AnTuTu v11
  5. GeekBench 6
  6. Specs
  7. Comments (6)

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 G3
  • Has 3 more cores
  • Supports 33% higher memory bandwidth (68.2 against 51.2 GB/s)
  • Better instruction set architecture
Pros of Apple A17 Pro
  • Higher GPU frequency (~57%)
  • 30% higher CPU clock speed (3780 vs 2910 MHz)
  • Has a smaller size transistor (3 versus 4 nm)
  • Performs 35% better in floating-point computations
  • Shows better (up to 4%) AnTuTu 11 score – 1467K vs 1409K

Benchmarks

Performance tests in popular benchmarks
SoC:
Tensor G3
vs
A17 Pro

AnTuTu 11

The AnTuTu Benchmark measures CPU, GPU, RAM, and I/O performance in different scenarios
Tensor G3
1409016
A17 Pro +4%
1467470
CPU 482403 374357
GPU 384611 463612
Memory 218603 184670
UX 323399 444831
Total score 1409016 1467470
Samples [33 tests] [21 tests]
AnTuTu results from iOS and Android are not directly comparable! They won't affect NanoReview scores.
Submit your AnTuTu result

GeekBench 6

The GeekBench test shows raw single-threaded and multithreaded CPU performance
Single-Core Score
Tensor G3
1722
A17 Pro +72%
2969
Multi-Core Score
Tensor G3
4458
A17 Pro +70%
7599
Asset compression 197.6 MB/sec 254.1 MB/sec
HTML 5 Browser 134.7 pages/sec 166 pages/sec
PDF Renderer 148.4 Mpixels/sec 180.7 Mpixels/sec
Image detection 82.1 images/sec 178.7 images/sec
HDR 142.1 Mpixels/sec 240.3 Mpixels/sec
Background blur 13.2 images/sec 26.7 images/sec
Photo processing 48.4 images/sec 77.8 images/sec
Ray tracing 5.34 Mpixels/sec 7.76 Mpixels/sec
Samples 11 tests 32 tests
Compute Score (GPU)
Tensor G3
8076
A17 Pro +247%
28057
Background blur 10.6 images/sec 55.5 images/sec
Face detection 6.85 images/sec 65.1 images/sec
Horizon detection 0.16 Gpixels/sec 0.8 Gpixels/sec
Edge detection 0.35 Gpixels/sec 1.08 Gpixels/sec
Gaussian blur 0.32 Gpixels/sec 1.11 Gpixels/sec
Feature matching 0.08 Gpixels/sec 0.43 Gpixels/sec
Stereo matching 11.9 Gpixels/sec 87.6 Gpixels/sec
Particle physics 1088 FPS 3313.9 FPS
API Vulkan Metal
Samples 9 tests 22 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
Tensor G3
982
A17 Pro +42%
1391
Score 982 1391
Stability 74% 71%
Graphics test 7 FPS 10 FPS
3DMark Solar Bay
Tensor G3
n/a
A17 Pro
6089
Score - 6089
Stability - 73%
Graphics test - 23 FPS
Section 1 - 29 FPS
Section 2 - 23 FPS
Section 3 - 16 FPS
3DMark Wild Life Extreme
Tensor G3
n/a
A17 Pro
3655
Score - 3655
Stability - 63%
Graphics test - 21 FPS

Recent User Tests

The latest benchmark tests that have been submitted by users
Google Tensor G3
DateBenchmarkResult
πŸ“˜ 2026-10-04 (Karlik)AnTuTu 121408420
πŸ“˜ 2026-08-28 (asfiii-malik)AnTuTu 111515235
πŸ“˜ 2026-08-28 (asfiii-baloch)AnTuTu 111515235
πŸ“˜ 2026-05-05 (s1mpick)AnTuTu 111338293
πŸ“˜ 2026-05-04 (reazz)AnTuTu 111356451
Add View all (49)
Apple A17 Pro
DateBenchmarkResult
πŸ“˜ 2026-10-01 (Zabza)AnTuTu 111996792
πŸ“˜ 2026-09-28 (Kuan.28)AnTuTu 111752198
πŸ“˜ 2026-09-23 (podsolnuh)AnTuTu 111357765
πŸ“˜ 2026-09-11 (Zabza)AnTuTu 111599694
πŸ“˜ 2026-09-10 (elio)AnTuTu 111445771
Add View all (71)

Specifications

Full list of technical specifications of Tensor G3 and A17 Pro

CPU

Architecture 1x 2.91 GHz – Cortex-X3
4x 2.37 GHz – Cortex-A715
4x 1.7 GHz – Cortex-A510		 2x 3.78 GHz – Everest
4x 2.11 GHz – Sawtooth
Cores 9 6
Frequency 2910 MHz 3780 MHz
Instruction set ARMv9-A ARMv8.6-A
L1 cache - 256 KB
L2 cache - 16 MB
Process 4 nanometers 3 nanometers
Transistor count - 19 billion
TDP (Sustained Power Limit) 6 W 8 W
Manufacturing Samsung TSMC

Graphics

GPU name Mali-G715 MP7 Apple A17 GPU
Architecture Valhall 4th gen Apple Bionic GPU
GPU frequency 890 MHz 1398 MHz
Pipelines 7 6
Shading units 128 128
Total shaders 896 768
FLOPS 1594.8 Gigaflops 2147.2 Gigaflops
Vulkan version 1.3 -
OpenCL version 2.0 -

AI Accelerator

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

Memory

Memory type LPDDR5X LPDDR5
Memory frequency 4266 MHz 3200 MHz
Bus 4x 16 Bit 4x 16 Bit
Max bandwidth 68.2 Gb/s 51.2 Gb/s
Max size 16 GB 8 GB

Multimedia (ISP)

Storage type UFS 3.1, UFS 4.0 NVMe
Max display resolution 3840 x 2160 1290 x 2796
Max camera resolution 1x 200MP, 2x 32MP 1x 48MP
Video capture 8K at 30FPS, 4K at 120FPS 4K at 60FPS
Video playback 8K at 30FPS, 4K at 120FPS 4K at 60FPS
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 5300i Snapdragon X70
4G support LTE Cat. 24 LTE Cat. 24
5G support Yes Yes
Download speed Up to 7300 Mbps Up to 7500 Mbps
Upload speed Up to 3700 Mbps Up to 3500 Mbps
Wi-Fi 7 6
Bluetooth 5.3 5.3
Navigation GPS, GLONASS, Beidou, Galileo, QZSS GPS, GLONASS, Beidou, Galileo, QZSS, NAVIC

Info

Announced October 2023 September 2023
Class Flagship Flagship
Model number GS301 (S5P9865) APL1V02
Official page Google Tensor G3 official site -
Successor - Google Tensor G5 - Apple A19 Pro

Cast your vote

So, which SoC would you choose?
243 (38.2%)
393 (61.8%)
Total votes: 636

Related Comparisons

1. Google Tensor G3 vs Google Tensor G2 (6)
2. Google Tensor G3 vs Qualcomm Snapdragon 8 Gen 2 (13)
3. Google Tensor G3 vs Qualcomm Snapdragon 8 Gen 3 (3)
4. Google Tensor G3 vs Google Tensor
5. Google Tensor G3 vs Qualcomm Snapdragon 8 Gen 1 (7)
6. Apple A17 Pro vs Apple M1 (iPad) (8)
7. Apple A17 Pro vs Apple A16 Bionic (19)
8. Apple A17 Pro vs Apple A18 Pro (19)
9. Apple A17 Pro vs Google Tensor G4
10. Apple A17 Pro vs Qualcomm Snapdragon 8 Gen 3 (141)
Compare other chipsets (190+)

Comments

Please give your opinion on the comparison of A17 Pro and Tensor G3, or ask any questions
🌐 Register your profile and become part of NanoReview community!
Avatar
dont at me October 18, 2023 at 11:58 AM
The A17 Pro's GPU is insane though. I sincerely hope game devs will hop on board with the metal API and bring actually cool games, not the bs that is mobile gaming, to the 15 Pro. It'd be so sick to have what's basically a handheld console with an amazing screen in your pocket that just so happens to also be your phone and have a great camera, that's how I view it anyway and it's also the only reason I'm even interested in it. I really hope it'll be utilized, everything else would be such a waste of that amazing GPU.
+5 Reply
Avatar
Google's Scam October 7, 2023 at 12:10 PM
This is a joke. Google selling 1000$ phone with a 4-year-old chip! Apple's 4-year-old A14 Bionic is better than this sh!t.
+25 Reply
Avatar
alex October 17, 2023 at 07:27 AM
The Tensor G3 has a much lower heat release than the A17 Pro.
+11 Reply
Avatar
Joe October 17, 2023 at 04:12 PM
alex, No, no. Google is selling a $ 1,000 phone with the most powerful artificial intelligence chip, arguably the best in the market and it boasts some of the best camera software in phones, ever. Samsung and Google are overall the best. However, with IOS, the iPhone could be considered just a beginner's phone.
+20 Reply
Avatar
AI AI AI November 6, 2023 at 03:56 PM
Joe, Sure, an AI chip is present. Then why don't all the AI features work in offline mode? :) I was satisfied with the Pixel 3 and 5 for several years. However, after purchasing the Pixel 8, it feels like a scam.
+10 Reply
Avatar
Satoru March 29, 2024 at 05:29 AM
AI AI AI, Do they really? It's just android's Google 1 (now available for all Android devices) that doesn't work offline. The Tensor G3's major advantage is its capacity to constantly adapt. These, combined with the fact that all other Android phones release two newer generations (like S25+UltraExtreme, etc.) before getting the same Android updates and features available on the Pixel, create a competitive edge. In fact, competing Android brands typically need one more phone generation to access features that the Pixel already possesses. Its performance might not reach the insane benchmarks of some "top tier" chips, but it doesn't need to β€” it isn't an overpriced gaming chip. Rather, it's the only chip constructed with AI and machine learning (hence the 6-7 year OS updates promise). When you couple that with the Titan M2 security, it triumphs over other Android devices, including the AI-rich Samsung S24 Ultra, which only boasts of about 3-4 AI features, all constantly bombarding the user. In contrast, the G3 applies an AI feature to most tasks with a quiet confidence. I would however, disagree with the comment that iOS makes iPhones "beginner phones". There's value in its closed ecosystem, as it offers top-level performance in every benchmark, with apps functioning near-perfectly. Decades of dedicated development and optimization contribute to this, extracting all potential from a single phone model, something that Android doesn't do. This can lead to excessive and unnecessary bloatware in top tier Android phones. In my opinion, the iPhone probably outperforms most top-tier Android phones in most categories.
+2 Reply
EnglishРусский