Snapdragon 888 vs Google Tensor

We compared two 8-core processors: Qualcomm Snapdragon 888 (with Adreno 660 graphics) and Google Tensor (Mali-G78 MP20). 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 Qualcomm Snapdragon 888
  • Performs 19% better in floating-point computations
  • Shows better (up to 12%) AnTuTu 11 score – 1135K vs 1010K
  • Better instruction set architecture

Benchmarks

Performance tests in popular benchmarks

AnTuTu 11

The AnTuTu Benchmark measures CPU, GPU, RAM, and I/O performance in different scenarios
1135878
1010025
CPU 406119 365546
GPU 233383 169803
Memory 188085 178763
UX 308291 295913
Total score 1135878 1010025
Samples [127 tests] [51 tests]

GeekBench 6

The GeekBench test shows raw single-threaded and multithreaded CPU performance
Single-Core Score
Multi-Core Score
Asset compression 156.9 MB/sec 159.8 MB/sec
HTML 5 Browser 86.4 pages/sec 97.5 pages/sec
PDF Renderer 104.6 Mpixels/sec 138.9 Mpixels/sec
Image detection 72.9 images/sec 83.4 images/sec
HDR 104.7 Mpixels/sec 119.7 Mpixels/sec
Background blur 12.1 images/sec 14.1 images/sec
Photo processing 26.5 images/sec 30.2 images/sec
Ray tracing 4.04 Mpixels/sec 3.44 Mpixels/sec
Samples 49 tests 26 tests
Compute Score (GPU)
8918
Background blur 7.75 images/sec 13.2 images/sec
Face detection 3.37 images/sec 7.31 images/sec
Horizon detection 0.22 Gpixels/sec 0.34 Gpixels/sec
Edge detection 0.28 Gpixels/sec 0.5 Gpixels/sec
Gaussian blur 0.32 Gpixels/sec 0.55 Gpixels/sec
Feature matching 0.03 Gpixels/sec 0.09 Gpixels/sec
Stereo matching 13.1 Gpixels/sec 39.4 Gpixels/sec
Particle physics 851.2 FPS 1368.9 FPS
API OpenCL Vulkan
Samples 19 tests 11 tests

3DMark

A cross-platform benchmark that assesses graphics performance in Vulkan (Metal)
3DMark Steel Nomad Light
Score 505 704
Stability 77% 63%
Graphics test 4 FPS 5 FPS
3DMark Wild Life Extreme
1899
Score 1508 1899
Stability 91% 59%
Graphics test 9 FPS 11 FPS

PCMark 3.0

Benchmark for evaluating overall system performance in everyday workloads
PCMark 3.0
13229
Web score 12232 7105
Video editing score 7638 5884
Data manipulation score 10494 10310
Photo editing score 29866 19072
Writing score 15219 15550
Phones tested 11 models 3 models

Gaming

Table of average FPS and graphics settings in mobile games
PUBG Mobile 83 FPS
[Ultra]
-
Call of Duty: Mobile 60 FPS
[Ultra]
-
Fortnite 26 FPS
[Ultra]
-
Shadowgun Legends 89 FPS
[Ultra]
-
World of Tanks Blitz 112 FPS
[Ultra]
-
Genshin Impact 58 FPS
[Ultra]
-
Mobile Legends: Bang Bang 60 FPS
[Ultra]
-
Device Asus Zenfone 8
1080 x 2400
-
We provide average results. FPS may differ, depending on game version, OS and other factors.

Recent User Tests

The latest benchmark tests that have been submitted by users
Qualcomm Snapdragon 888
DateBenchmarkResult
πŸ“˜ 2026-09-12 (Zha)AnTuTu 12976692
πŸ“˜ 2026-09-10 (Aladdin)AnTuTu 121326171
πŸ“˜ 2026-09-09 (Aladdin)AnTuTu 121326171
πŸ“˜ 2026-09-02 -> NotezitGeekbench 6 CPU1412 / 3348
πŸ“˜ 2026-08-18 (Soul)AnTuTu 111210861
Google Tensor
DateBenchmarkResult
πŸ“˜ 2026-09-29 (rulz_103)AnTuTu 121225085
πŸ“˜ 2026-09-26 -> Islom_1180AnTuTu 121235821
πŸ“˜ 2026-09-13 (Vladi♂️Slave♂️)AnTuTu 121281085
πŸ“˜ 2026-08-19 (duwi)AnTuTu 11774697
πŸ“˜ 2026-08-17 -> Paranoid _AntostAnTuTu 11930801

Specifications

Full list of technical specifications of Snapdragon 888 and Google Tensor

CPU

Architecture 1x 2.84 GHz – Kryo 680 Prime (Cortex-X1)
3x 2.42 GHz – Kryo 680 Gold (Cortex-A78)
4x 1.8 GHz – Kryo 680 Silver (Cortex-A55)
2x 2.8 GHz – Cortex-X1
2x 2.25 GHz – Cortex A76
4x 1.8 GHz – Cortex A55
Cores 8 8
Frequency 2840 MHz 2800 MHz
Instruction set ARMv8.4-A ARMv8.2-A
L1 cache 512 KB -
L2 cache 1 MB 3 MB
L3 cache 4 MB 4 MB
Process 5 nanometers 5 nanometers
Transistor count 10.3 billion -
TDP (Sustained Power Limit) 8 W 5.6 W
Manufacturing Samsung Samsung

Graphics

GPU name Adreno 660 Mali-G78 MP20
Architecture Adreno 600 Valhall 2nd gen
GPU frequency 840 MHz 850 MHz
Pipelines 3 20
Shading units 256 32
Total shaders 768 640
FLOPS 1290.2 Gigaflops 1085.4 Gigaflops
Vulkan version 1.1 1.3
OpenCL version 2.0 2.0
DirectX version 12.1 -

AI Accelerator

Neural processor (NPU) Hexagon 780 Tensor Processing Unit

Memory

Memory type LPDDR5 LPDDR5
Memory frequency 3200 MHz 3200 MHz
Bus 4x 16 Bit 4x 16 Bit
Max bandwidth 51.2 Gb/s 51.2 Gb/s
Max size 24 GB 12 GB

Multimedia (ISP)

Storage type UFS 3.0, UFS 3.1 UFS 3.1
Max display resolution 3840 x 2160 3840 x 2160
Max camera resolution 1x 200MP, 2x 25MP 1x 200MP, 2x 32MP
Video capture 8K at 30FPS, 4K at 120FPS 4K at 60FPS
Video playback 8K at 30FPS 4K at 60FPS
Video codecs - H.264
- H.265
- VP8
- VP9
- H.264
- H.265
- AV1
- VP9
Audio codecs - AAC
- AIFF
- CAF
- MP3
- MP4
- WAV
- AAC
- AIFF
- CAF
- MP3
- MP4
- WAV

Connectivity

Modem X60 Exynos 5300b
4G support LTE Cat. 22 LTE Cat. 24
5G support Yes Yes
Download speed Up to 7500 Mbps Up to 7350 Mbps
Upload speed Up to 3000 Mbps Up to 3670 Mbps
Wi-Fi 6 6
Bluetooth 5.2 5.2
Navigation GPS, GLONASS, Beidou, Galileo, QZSS, SBAS, NAVIC GPS, GLONASS, Beidou, Galileo, QZSS

Info

Announced December 2020 October 2021
Class Flagship Flagship
Model number SM8350 GS101 (S5P9845)
Official page Qualcomm Snapdragon 888 official site Google Tensor official site
Successor - Qualcomm Snapdragon 8 Elite Gen 5 - Google Tensor G5

Cast your vote

So, which SoC would you choose?
2569 (35.4%)
4683 (64.6%)
Total votes: 7252

Comments

Please give your opinion on the comparison of Google Tensor and Snapdragon 888, or ask any questions
🌐 Register your profile and become part of NanoReview community!
Avatar
Ronald Campos March 11, 2025 at 05:26 AM
Maybe the Google Tensor's score is outdated. I'm pretty sure it'll outperform the Snapdragon 888 5G for sure when retesting it with the newer, updated Linux Kernel 6.1.99 version.
+13 Reply
Avatar
Guna(iqoo 3 user) February 20, 2025 at 09:46 AM
Even sd 865 is better than tensor in gaming. They are similar in normal usage but I prefer 865. 865 is far better. I have pixel 6a and iqoo 3.
+9 Reply
Avatar
Oleg March 28, 2024 at 04:28 PM
Please retest Snapdragon 888 in all benchmarks, as your data appears to be outdated. My phone is powered by Snapdragon 888, and it scored 1520/3927 in Geekbench 6.
+58 Reply
Avatar
Anon March 17, 2023 at 09:17 AM
OnePlus 9 Pro or Pixel 6 Pro? I'm not a gamer, but I want to future-proof for stability and reliability of the OS.
+199 Reply
Avatar
Neeraj April 6, 2023 at 01:24 AM
I think you should go with pixel 6 pro.
+88 Reply
Avatar
Ansh December 31, 2024 at 01:15 PM
Pixel 6 pro is future proof due to software and hardware optimization
+5 Reply
Avatar
LAX February 17, 2023 at 08:58 PM
Sd888 is better with a good CPU set-up well optimized for huge processes like games it's actually a beast as compared to the Google sensor. Tensor is good but needs more time to catch up with the king of chipsets SNAPDRAGON.
+51 Reply
Avatar
Anonymous User December 24, 2023 at 08:01 PM
I believe the Tensor outperforms the Snapdragon 888. Nevertheless, the Snapdragon 8 Gen surpasses the Tensor. However, if you were to compare the Snapdragon 888 with the Tensor G2, then you might be right.
+19 Reply
Avatar
Hubert March 12, 2022 at 11:46 AM
The Snapdragon 888 has no AV1 Codec.
+107 Reply
Avatar
Sergey March 12, 2022 at 12:31 PM
Sorry for that, we fixed the error.
+51 Reply
Avatar
Daviddoff February 5, 2022 at 02:12 PM
Isn't the tensor supposed to have learning capabilities that the 888 doesn't have or is that just made up? I thought the tensor is supposed to learn your habits and mold to each user in time?
+181 Reply
Avatar
HHM October 28, 2021 at 09:19 AM
Battery life in Tensor is not good, but for the 1st gen, let's say it's not bad. Maybe, in the future, Google will optimise Tensor so it will be able to compete with Apple A processors.
+100 Reply