Google Tensor Benchmark, Test and specs

Last updated:
The Google Tensor has 8 cores with 8 threads and is based on the 1. gen of the Google Tensor series. The processor was released in Q4/2021. The Google Tensor scores 1,043 points in the Geekbench 5 single-core benchmark. In the Geekbench 5 multi-core benchmark, the result is 2,915 points.
Google Tensor

At a glance

Name: Google Tensor
Family: Google Tensor (4)
CPU group: Google Tensor (1)
Architecture: G1
Technology: 5 nm
Segment: Smartphone / Tablet
Generation: 1
Predecessor: --
Successor: Google Tensor G2

CPU Cores and Base Frequency

The Google Tensor has 8 CPU cores and can calculate 8 threads in parallel. The clock frequency of the Google Tensor is 2.80 GHz. The number of CPU cores greatly affects the speed of the processor and is an important performance indicator.

CPU Cores / Threads: 8 / 8
Core architecture: hybrid (Prime / big.LITTLE)
A-Core: 2x Cortex-X1
B-Core: 2x Cortex-A76
C-Core: 4x Cortex-A55
Hyperthreading / SMT: No
Overclocking: No
A-Core Frequency: 2.80 GHz
B-Core Frequency: 2.25 GHz
C-Core Frequency: 1.80 GHz

NPU AI performance

The performance values of the processor's AI unit. The isolated NPU performance is specified here, the total AI performance (NPU+CPU+iGPU) can be higher. Processors with support for artificial intelligence (AI) and machine learning (ML) can process many calculations, especially audio, image and video processing, much faster than classic processors.

AI hardware: Google Tensor AI
AI specifications: Google Edge TPU @ 1.6 TOPS
NPU + CPU + iGPU: --

Integrated graphics (iGPU)

The Google Tensor has integrated graphics, called iGPU for short. Specifically, the Google Tensor uses the ARM Mali-G78 MP20, which has 320 texture shaders and 20 execution units. The iGPU uses the system's main memory as graphics memory and sits on the processor's die.

GPU name: ARM Mali-G78 MP20
GPU frequency: 0.76 GHz
GPU (Turbo): No turbo
Compute units: 20
Shader: 320
Hardware Raytracing: No
Release date: Q4/2021
Max. displays: 1
Generation: Vallhall 2
Direct X: 12
Technology: 5 nm
Max. GPU Memory: --
Frame Generation: No

Hardware codec support

A photo or video codec that is accelerated in hardware can greatly accelerate the working speed of a processor and extend the battery life of notebooks or smartphones when playing videos.

h265 / HEVC (8 bit): Decode / Encode
h265 / HEVC (10 bit): Decode / Encode
h264: Decode / Encode
VP8: Decode / Encode
VP9: Decode / Encode
AV1: Decode
AVC: Decode / Encode
VC-1: Decode / Encode
JPEG: Decode / Encode

Memory & PCIe

The processor can use up to 12 GB memory in 2 (Dual Channel) memory channels. The maximum memory bandwidth is 53.0 GB/s. The memory type as well as the amount of memory can greatly affect the speed of the system.

Memory type: Memory bandwidth:
LPDDR5-5500
53.0 GB/s
Max. Memory: 12 GB
Memory channels: 2 (Dual Channel)
ECC: No
PCIe:
PCIe Bandwidth: --

Thermal Management

The thermal design power (TDP for short) of the processor is 10 W. The TDP specifies the necessary cooling solution that is required to cool the processor sufficiently. The TDP usually gives a rough idea of the actual power consumption of the CPU.

TDP (PL1 / PBP): 10 W
TDP (PL2): --
TDP up: --
TDP down: --
Tjunction max.: --

Technical details

The Google Tensor is made in 5 nm. The smaller the manufacturing process of a CPU, the more modern and energy-efficient it is. Overall, the processor has 8.00 MB cache. A large cache can greatly speed up the processor's speed in some cases such as games.

Technology: 5 nm
Chip design:
Socket: --
L2-Cache: 8.00 MB
L3-Cache: --
AES-NI: No
Operating systems: Android
Virtualization: None
Instruction set (ISA): Armv8-A (64 bit)
ISA extensions: --
Release date: Q4/2021
Release price: --
Part Number: --
Documents: --

Rate this processor

Here you can rate this processor to help other visitors make their purchase decision. The average rating of this processor is 4.4 stars (55 ratings). Rate now:

Benchmark results

Verified Benchmark results
The benchmark results for the Google Tensor have been carefully checked by us. We only publish benchmark results that have been created by us or that have been submitted by a visitor and then checked by a team member. All results are based on and fullfill our benchmark guidelines.

Screenshots:

Geekbench 6 (Single-Core)

Geekbench 6 is a partial load benchmark for modern computers, notebooks and smartphones. In the single-core test, only the fastest CPU core is measured. The test run simulates the performance in practice.
Qualcomm Snapdragon 8+ Gen 1 Qualcomm Snapdragon 8+ Gen 1
8C 8T @ 3.20 GHz
1657
MediaTek Dimensity 9000+ MediaTek Dimensity 9000+
8C 8T @ 3.20 GHz
1521
Qualcomm Snapdragon 7+ Gen 2 Qualcomm Snapdragon 7+ Gen 2
8C 8T @ 2.91 GHz
1496
Google Tensor Google Tensor
8C 8T @ 2.80 GHz
1494
Samsung Exynos 2200 Samsung Exynos 2200
8C 8T @ 2.80 GHz
1428
Google Tensor G2 Google Tensor G2
8C 8T @ 2.85 GHz
1426
Apple A12Z Bionic Apple A12Z Bionic
8C 8T @ 2.49 GHz
1351

Geekbench 6 (Multi-Core)

The practical Geekbench 6 multi-core benchmark tests the system's performance under partial load. The processor's maximum power consumption is far from being used up.
Qualcomm Snapdragon 8+ Gen 1 Qualcomm Snapdragon 8+ Gen 1
8C 8T @ 3.20 GHz
4231
MediaTek Dimensity 9000+ MediaTek Dimensity 9000+
8C 8T @ 3.20 GHz
4223
Qualcomm Snapdragon 7+ Gen 2 Qualcomm Snapdragon 7+ Gen 2
8C 8T @ 2.91 GHz
4189
Google Tensor Google Tensor
8C 8T @ 2.80 GHz
3639
MediaTek Dimensity 8020 MediaTek Dimensity 8020
8C 8T @ 2.60 GHz
3631
Apple A13 Bionic Apple A13 Bionic
6C 6T @ 2.65 GHz
3599
Samsung Exynos 2200 Samsung Exynos 2200
8C 8T @ 2.80 GHz
3528

Geekbench 5, 64bit (Single-Core)

Geekbench 5 is a cross plattform benchmark that heavily uses the systems memory. A fast memory will push the result a lot. The single-core test only uses one CPU core, the amount of cores or hyperthreading ability doesn't count.
HiSilicon Kirin 9000 HiSilicon Kirin 9000
8C 8T @ 3.13 GHz
1063
HiSilicon Kirin 9000E HiSilicon Kirin 9000E
8C 8T @ 3.13 GHz
1063
Samsung Exynos 1580 Samsung Exynos 1580
8C 8T @ 2.91 GHzNot verified
1046
Google Tensor Google Tensor
8C 8T @ 2.80 GHz
1043
Samsung Exynos 2100 Samsung Exynos 2100
8C 8T @ 2.90 GHz
1011
Qualcomm Snapdragon 870 Qualcomm Snapdragon 870
8C 8T @ 3.20 GHz
996
MediaTek Dimensity 8200 MediaTek Dimensity 8200
8C 8T @ 3.10 GHz
987

Geekbench 5, 64bit (Multi-Core)

Geekbench 5 is a cross plattform benchmark that heavily uses the systems memory. A fast memory will push the result a lot. The multi-core test involves all CPU cores and taks a big advantage of hyperthreading.
Qualcomm Snapdragon 4 Gen 1 Qualcomm Snapdragon 4 Gen 1
8C 8T @ 2.00 GHz
2979
MediaTek Dimensity 1000 MediaTek Dimensity 1000
8C 8T @ 2.60 GHz
2976
Qualcomm Snapdragon 7 Gen 1 Qualcomm Snapdragon 7 Gen 1
6C 6T @ 2.40 GHz
2932
Google Tensor Google Tensor
8C 8T @ 2.80 GHz
2915
Qualcomm Snapdragon 7s Gen 2 Qualcomm Snapdragon 7s Gen 2
8C 8T @ 0.70 GHz
2874
MediaTek Dimensity 8050 MediaTek Dimensity 8050
8C 8T @ 3.00 GHz
2861
Qualcomm Snapdragon 778G+ Qualcomm Snapdragon 778G+
8C 8T @ 2.50 GHz
2856

iGPU - FP32 Performance (Single-precision GFLOPS)

The theoretical computing performance of the internal graphics unit of the processor with simple accuracy (32 bit) in GFLOPS. GFLOPS indicates how many billion floating point operations the iGPU can perform per second.
Apple A17 Pro Apple A17 Pro
Apple A17 Pro (6 GPU Cores) @ 1.40 GHz
2147
HiSilicon Kirin 9000E HiSilicon Kirin 9000E
ARM Mali-G78 MP22 @ 0.76 GHz
2137
Qualcomm Snapdragon 8 Elite for Galaxy Qualcomm Snapdragon 8 Elite for Galaxy
Qualcomm Adreno 830 @ 1.20 GHz
1950
Google Tensor Google Tensor
ARM Mali-G78 MP20 @ 0.76 GHz
1943
Qualcomm Snapdragon 8cx Qualcomm Snapdragon 8cx
Qualcomm Adreno 680 @ 0.59 GHz
1843
Qualcomm Snapdragon 8 Gen 2 for Galaxy Qualcomm Snapdragon 8 Gen 2 for Galaxy
Qualcomm Adreno 740 @ 0.72 GHz
1840
Qualcomm Snapdragon 8s Gen 3 Qualcomm Snapdragon 8s Gen 3
Qualcomm Adreno 740 @ 0.72 GHz
1840

AnTuTu 9 Benchmark

The AnTuTu 9 benchmark is very well suited to measuring the performance of a smartphone. AnTuTu 9 is quite heavy on 3D graphics and can now also use the "Metal" graphics interface. In AnTuTu, memory and UX (user experience) are also tested by simulating browser and app usage. AnTuTu version 9 can compare any ARM CPU running on Android or iOS. Devices may not be directly comparable when benchmarked on different operating systems.

In the AnTuTu 9 benchmark, the single-core performance of a processor is only slightly weighted. The rating is made up of the multi-core performance of the processor, the speed of the working memory, and the performance of the internal graphics.
Qualcomm Snapdragon 870 Qualcomm Snapdragon 870
8C 8T @ 3.20 GHz
727650
Samsung Exynos 2100 Samsung Exynos 2100
8C 8T @ 2.90 GHz
724660
Qualcomm Snapdragon 865+ Qualcomm Snapdragon 865+
8C 8T @ 3.10 GHz
716498
Google Tensor Google Tensor
8C 8T @ 2.80 GHz
691770
MediaTek Dimensity 1300 MediaTek Dimensity 1300
8C 8T @ 3.00 GHz
689042
MediaTek Dimensity 1200 MediaTek Dimensity 1200
8C 8T @ 3.00 GHz
669042
Qualcomm Snapdragon 865 Qualcomm Snapdragon 865
8C 8T @ 2.84 GHz
668494

AnTuTu 8 Benchmark

The AnTuTu 8 Benchmark measures the performance of a SoC. AnTuTu benchmarks the CPU, GPU, Memory as well as the UX (User Experience) by simulating browser and app usage. AnTuTu can benchmark any ARM CPU that runs under Android or iOS. Devices may not be directly compareable if the benchmark has been performed under different operating systems.

In the AnTuTu 8 benchmark, the single-core performance of a processor is only slightly weighted. The evaluation consists of the multi-core performance of the processor, the speed of the RAM and the performance of the internal graphics.
Apple A14 Bionic Apple A14 Bionic
6C 6T @ 3.00 GHz
628047
MediaTek Dimensity 1200 MediaTek Dimensity 1200
8C 8T @ 3.00 GHz
627817
Qualcomm Snapdragon 865+ Qualcomm Snapdragon 865+
8C 8T @ 3.10 GHz
616032
Google Tensor Google Tensor
8C 8T @ 2.80 GHz
612494
Samsung Exynos 2100 Samsung Exynos 2100
8C 8T @ 2.90 GHz
602990
Qualcomm Snapdragon 865 Qualcomm Snapdragon 865
8C 8T @ 2.84 GHz
598103
Qualcomm Snapdragon 860 Qualcomm Snapdragon 860
8C 8T @ 2.96 GHz
560128

AI performance (NPU)

The performance values of the processor's AI unit. The isolated NPU performance is given here, the total AI performance (NPU+CPU+iGPU) can be higher.

Processors with the support of artificial intelligence (AI) and machine learning (ML) can process many calculations, especially audio, image and video processing, much faster than classic processors. The performance is given in the number (trillions) of arithmetic operations per second (TOPS).
Qualcomm Snapdragon 665 Qualcomm Snapdragon 665
8C 8T @ 2.00 GHz
3
Qualcomm Snapdragon 670 Qualcomm Snapdragon 670
8C 8T @ 2.00 GHz
3
Qualcomm Snapdragon 8cx Qualcomm Snapdragon 8cx
8C 8T @ 2.84 GHz
3
Google Tensor Google Tensor
8C 8T @ 2.80 GHz
1.6
MediaTek Helio G99 MediaTek Helio G99
8C 8T @ 2.20 GHz
1
Apple A11 Bionic Apple A11 Bionic
6C 6T @ 2.39 GHz
0.6
MediaTek MT8183 MediaTek MT8183
8C 8T @ 2.00 GHz
0.5

Benchmarks


Description of the processor

The Google Tensor is a 64-bit system-on-a-chip (SOC) processor developed by the American company Google. It was released in the fourth quarter of 2021 and was used in Googles own smartphones Google Pixel 6, Google Pixel 6 Pro and Google Pixel 6a. The Google Tensor is the first generation of Tensor processors and is manufactured with a structure width of 5 nanometers. With the Google Tensor G2, the successor to the first generation came in 2022, which will be installed in the Google Pixel 7.

The Google Tensor is based on a hybrid Prime big.LITTLE core architecture and has a total of eight processor cores. These are divided into 2 prime cores, 2 performance cores and 4 efficiency cores. The two prime cores clock at up to 2.80 gigahertz and are based on an ARM Cortex-X1 core. The two performance cores are based on the ARM Cortex-A76 and clock at up to 2.25 gigahertz. The four efficiency cores, which are used when no computing power is needed to extend the battery life of the smartphone, are based on the ARM Cortex-A55 and clock at a maximum of 1.80 gigahertz.

With the Google Tensor AI (Google Edge TPU with 1.6 TOPS performance), special hardware is built into the Google Tensor, which supports the calculation of AI or ML in hardware.

The internal graphics unit in the Google Tensor is the ARM Mali-G78 with 20 execution units. This iGPU has a total of 320 shader units and clocks at up to 760 megahertz; the graphics unit does not have a turbo mode. It achieves an FP32 computing power of 1943 GigaFLOPS, with single precision. The ARM Mail-G78 is manufactured with a structure width of 5 nanometers and comes from the Valhall 2 generation.

The Google Tensor G1 was equipped with up to 12 gigabytes of LPDDR5-5500 memory and has 2 memory channels.



Popular comparisons

1. Google Tensor Qualcomm Snapdragon 888 Google Tensor vs Qualcomm Snapdragon 888
2. Qualcomm Snapdragon 8 Gen 1 Google Tensor Qualcomm Snapdragon 8 Gen 1 vs Google Tensor
3. Google Tensor Google Tensor G2 Google Tensor vs Google Tensor G2
4. Google Tensor Qualcomm Snapdragon 865 Google Tensor vs Qualcomm Snapdragon 865
5. Google Tensor Qualcomm Snapdragon 695 5G Google Tensor vs Qualcomm Snapdragon 695 5G
6. Google Tensor Qualcomm Snapdragon 855 Google Tensor vs Qualcomm Snapdragon 855
7. Google Tensor G3 Google Tensor Google Tensor G3 vs Google Tensor
8. Qualcomm Snapdragon 8 Gen 2 Google Tensor Qualcomm Snapdragon 8 Gen 2 vs Google Tensor
9. Google Tensor Qualcomm Snapdragon 870 Google Tensor vs Qualcomm Snapdragon 870
10. Apple M1 Google Tensor Apple M1 vs Google Tensor


back to index