Google Tensor G3 vs AMD EPYC 9754

Last updated:

CPU comparison with benchmarks


Google Tensor G3 CPU1 vs CPU2 AMD EPYC 9754
Google Tensor G3 AMD EPYC 9754

CPU comparison

Google Tensor G3 or AMD EPYC 9754 - which processor is faster? In this comparison we look at the differences and analyze which of these two CPUs is better. We compare the technical data and benchmark results.

The Google Tensor G3 has 8 cores with 8 threads and clocks with a maximum frequency of 2.91 GHz. Up to 12 GB of memory is supported in 2 memory channels. The Google Tensor G3 was released in Q3/2023.

The AMD EPYC 9754 has 128 cores with 256 threads and clocks with a maximum frequency of 3.10 GHz. The CPU supports up to 6144 GB of memory in 12 memory channels. The AMD EPYC 9754 was released in Q2/2023.
Google Tensor (3) Family AMD EPYC (129)
Google Tensor G3 (1) CPU group AMD EPYC 8004/9004 (15)
3 Generation 4
G3 Architecture Bergamo/Siena (Zen 4c)
Mobile Segment Desktop / Server
Google Tensor Predecessor --
-- Successor --

CPU Cores and Base Frequency

The Google Tensor G3 is a 8 core processor with a clock frequency of 2.91 GHz. The AMD EPYC 9754 has 128 CPU cores with a clock frequency of 2.25 GHz (3.10 GHz).

Google Tensor G3 Characteristic AMD EPYC 9754
8 Cores 128
8 Threads 256
hybrid (Prime / big.LITTLE) Core architecture normal
No Hyperthreading Yes
No Overclocking ? No
2.91 GHz
1x Cortex-X3
A-Core 2.25 GHz (3.10 GHz)
128x Zen 4c
2.37 GHz
4x Cortex-A715
B-Core --
1.70 GHz
4x Cortex-A510
C-Core --

Artificial Intelligence and Machine Learning

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. Algorithms for ML improve their performance the more data they have collected via software. ML tasks can be processed up to 10,000 times faster than with a classic processor.

Google Tensor G3 Characteristic AMD EPYC 9754
Google Tensor AI AI hardware --
Google Edge TPU AI specifications --

Internal Graphics

The integrated graphics unit of a processor is not only responsible for the pure image output on the system, but can also significantly increase the efficiency of the system with the support of modern video codecs.

ARM Immortalis-G715 MP10 GPU no iGPU
0.89 GHz GPU frequency --
-- GPU (Turbo) --
Vallhall GPU Generation --
4 nm Technology
0 Max. displays
10 Compute units --
-- Shader --
No Hardware Raytracing No
No Frame Generation No
-- Max. GPU Memory --
12 DirectX Version --

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.

ARM Immortalis-G715 MP10 GPU no iGPU
Decode / Encode Codec h265 / HEVC (8 bit) No
Decode / Encode Codec h265 / HEVC (10 bit) No
Decode / Encode Codec h264 No
Decode / Encode Codec VP9 No
Decode / Encode Codec VP8 No
Decode / Encode Codec AV1 No
Decode / Encode Codec AVC No
Decode / Encode Codec VC-1 No
Decode / Encode Codec JPEG No

Memory & PCIe

The Google Tensor G3 supports a maximum of 12 GB of memory in 2 memory channels. The AMD EPYC 9754 can connect up to 6144 GB of memory in 12 memory channels.

Google Tensor G3 Characteristic AMD EPYC 9754
LPDDR5-5500 Memory DDR5-4800
12 GB Max. Memory 6144 GB
2 (Dual Channel) Memory channels 12
53.0 GB/s Max. Bandwidth 460.8 GB/s
No ECC Yes
-- L2 Cache --
-- L3 Cache 256.00 MB
-- PCIe version 5.0
-- PCIe lanes 128
-- PCIe Bandwidth 504.1 GB/s

Thermal Management

The TDP (Thermal Design Power) of a processor specifies the required cooling solution. The Google Tensor G3 has a TDP of 10 W, that of the AMD EPYC 9754 is 360 W.

Google Tensor G3 Characteristic AMD EPYC 9754
10 W TDP (PL1 / PBP) 360 W
-- TDP (PL2) --
-- TDP up 400 W
-- TDP down 320 W
-- Tjunction max. 100 °C

Technical details

The Google Tensor G3 has a 0.00 MB cache, while the AMD EPYC 9754 cache has a total of 256.00 MB.

Google Tensor G3 Characteristic AMD EPYC 9754
4 nm Technology 5 nm
Chiplet Chip design Chiplet
Armv9-A (64 bit) Instruction set (ISA) x86-64 (64 bit)
-- ISA extensions SSE4.2, AVX2, AVX-512, BFLOAT16, VNNI
-- Socket SP5
None Virtualization AMD-V, SVM
No AES-NI Yes
Android Operating systems Windows Server 2022, Linux, Windows 11
Q3/2023 Release date Q2/2023
-- Release price 11900 $
show more data show more data


Rate these processors

Here you can rate the Google Tensor G3 to help other visitors make their purchasing decisions. The average rating is 3.6 stars (28 ratings). Rate now:
Here you can rate the AMD EPYC 9754 to help other visitors make their purchasing decisions. The average rating is 4.7 stars (14 ratings). Rate now:


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.
Google Tensor G3 Google Tensor G3
8C 8T @ 2.91 GHz
1267 (100%)
AMD EPYC 9754 AMD EPYC 9754
128C 256T @ 3.10 GHz
0 (0%)

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.
Google Tensor G3 Google Tensor G3
8C 8T @ 2.91 GHz
3631 (100%)
AMD EPYC 9754 AMD EPYC 9754
128C 256T @ 3.10 GHz
0 (0%)

Geekbench 6 (Single-Core)

Geekbench 6 is a benchmark for modern computers, notebooks and smartphones. What is new is an optimized utilization of newer CPU architectures, e.g. based on the big.LITTLE concept and combining CPU cores of different sizes. The single-core benchmark only evaluates the performance of the fastest CPU core, the number of CPU cores in a processor is irrelevant here.
Google Tensor G3 Google Tensor G3
8C 8T @ 2.91 GHz
1759 (100%)
AMD EPYC 9754 AMD EPYC 9754
128C 256T @ 3.10 GHz
0 (0%)

Geekbench 6 (Multi-Core)

Geekbench 6 is a benchmark for modern computers, notebooks and smartphones. What is new is an optimized utilization of newer CPU architectures, e.g. based on the big.LITTLE concept and combining CPU cores of different sizes. The multi-core benchmark evaluates the performance of all of the processor's CPU cores. Virtual thread improvements such as AMD SMT or Intel's Hyper-Threading have a positive impact on the benchmark result.
Google Tensor G3 Google Tensor G3
8C 8T @ 2.91 GHz
4533 (100%)
AMD EPYC 9754 AMD EPYC 9754
128C 256T @ 3.10 GHz
0 (0%)

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.
Google Tensor G3 Google Tensor G3
ARM Immortalis-G715 MP10 @ 0.89 GHz
1 (100%)
AMD EPYC 9754 AMD EPYC 9754
@ 0.00 GHz
0 (0%)

AnTuTu 10 Benchmark

The AnTuTu 10 benchmark is one of the best-known benchmarks for mobile processors, which is now available in version 10. There is a version for Android-based smartphones and tablets, as well as a version for Apple mobile devices, i.e. iPhones and iPads.

The Antutu 10 benchmark has 3 phases. In the first phase, the device's RAM is tested, in phase 2 the graphics are tested and in the final phase the entire device is pushed to its performance limits by rendering 3D graphics.

Antutu 10 is therefore ideal for comparing the performance of different devices.
Google Tensor G3 Google Tensor G3
8C 8T @ 2.91 GHz
1106280 (100%)
AMD EPYC 9754 AMD EPYC 9754
128C 256T @ 2.25 GHz
0 (0%)

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.
Google Tensor G3 Google Tensor G3
8C 8T @ 2.91 GHz
894587 (100%)
AMD EPYC 9754 AMD EPYC 9754
128C 256T @ 2.25 GHz
0 (0%)

Devices using this processor

Google Tensor G3 AMD EPYC 9754
Google Pixel 8
Google Pixel 8 Pro
Unknown

Popular comparisons containing this CPUs

1. Qualcomm Snapdragon 8 Gen 2Google Tensor G3 Qualcomm Snapdragon 8 Gen 2 vs Google Tensor G3
2. Google Tensor G3Google Tensor G2 Google Tensor G3 vs Google Tensor G2
3. Qualcomm Snapdragon 8 Gen 3Google Tensor G3 Qualcomm Snapdragon 8 Gen 3 vs Google Tensor G3
4. Apple A17 ProGoogle Tensor G3 Apple A17 Pro vs Google Tensor G3
5. Qualcomm Snapdragon 8 Gen 1Google Tensor G3 Qualcomm Snapdragon 8 Gen 1 vs Google Tensor G3
6. Google Tensor G3Google Tensor Google Tensor G3 vs Google Tensor
7. Apple A16 BionicGoogle Tensor G3 Apple A16 Bionic vs Google Tensor G3
8. Qualcomm Snapdragon 888Google Tensor G3 Qualcomm Snapdragon 888 vs Google Tensor G3
9. Qualcomm Snapdragon 8 Gen 2 for GalaxyGoogle Tensor G3 Qualcomm Snapdragon 8 Gen 2 for Galaxy vs Google Tensor G3
10. Google Tensor G3Qualcomm Snapdragon 8cx Gen 3 Google Tensor G3 vs Qualcomm Snapdragon 8cx Gen 3


back to index