Tensor G5 vs Google Tensor
We compared two 8-core processors: Google Tensor G5 (with PowerVR DXT-48-1536 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
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
Pros of Google Tensor G5
- Has 4 MB larger L3 cache size than the Tensor
- Has a smaller size transistor (3 versus 5 nm)
- Performs 56% better in floating-point computations
- Shows significantly better (up to 53%) AnTuTu 11 score β 1549K vs 1010K
- 35% higher CPU clock speed (3780 vs 2800 MHz)
- Higher GPU frequency (~29%)
- Supports 33% higher memory bandwidth (68.2 against 51.2 GB/s)
- Better instruction set architecture
Benchmarks
Performance tests in popular benchmarksAnTuTu 11
| CPU | 734567 | 365546 |
| GPU | 183228 | 169803 |
| Memory | 222379 | 178763 |
| UX | 409624 | 295913 |
| Total score | 1549798 | 1010025 |
| Samples | [16 tests] | [51 tests] |
GeekBench 6
Single-Core Score
Multi-Core Score
| Asset compression | 281.8 MB/sec | 159.8 MB/sec |
| HTML 5 Browser | 162.5 pages/sec | 97.5 pages/sec |
| PDF Renderer | 207.4 Mpixels/sec | 138.9 Mpixels/sec |
| Image detection | 149.5 images/sec | 83.4 images/sec |
| HDR | 151.4 Mpixels/sec | 119.7 Mpixels/sec |
| Background blur | 24.4 images/sec | 14.1 images/sec |
| Photo processing | 46.5 images/sec | 30.2 images/sec |
| Ray tracing | 6.79 Mpixels/sec | 3.44 Mpixels/sec |
| Samples | 8 tests | 26 tests |
Compute Score (GPU)
| Background blur | 4.77 images/sec | 13.2 images/sec |
| Face detection | 2.41 images/sec | 7.31 images/sec |
| Horizon detection | 0.17 Gpixels/sec | 0.34 Gpixels/sec |
| Edge detection | 0.27 Gpixels/sec | 0.5 Gpixels/sec |
| Gaussian blur | 0.2 Gpixels/sec | 0.55 Gpixels/sec |
| Feature matching | 0.05 Gpixels/sec | 0.09 Gpixels/sec |
| Stereo matching | 12.6 Gpixels/sec | 39.4 Gpixels/sec |
| Particle physics | 634.9 FPS | 1368.9 FPS |
| API | Vulkan | Vulkan |
| Samples | 7 tests | 11 tests |
3DMark
3DMark Steel Nomad Light
| Score | 1012 | 704 |
| Stability | 96% | 63% |
| Graphics test | 7 FPS | 5 FPS |
3DMark Wild Life Extreme
| Score | - | 1899 |
| Stability | - | 59% |
| Graphics test | - | 11 FPS |
PCMark 3.0
PCMark 3.0
| Web score | 17237 | 7105 |
| Video editing score | 7203 | 5884 |
| Data manipulation score | 14036 | 10310 |
| Photo editing score | 24584 | 19072 |
| Writing score | 20213 | 15550 |
| Phones tested | 4 models | 3 models |
Recent User Tests
Google Tensor G5
| Date | Benchmark | Result |
|---|---|---|
| π 2026-09-25 (Dolbaeb) | AnTuTu 12 | 1549159 |
| π 2026-08-22 -> Pixeluser | AnTuTu 11 | 1536042 |
| π 2026-08-02 -> bang | AnTuTu 11 | 1719659 |
| π 2026-08-02 -> bang | CPU Throttle | 74% |
| π 2026-07-31 -> bang | Geekbench 6 GPU | 4116 |
Google Tensor
| Date | Benchmark | Result |
|---|---|---|
| π 2026-09-13 (VladiβοΈSlaveβοΈ) | AnTuTu 12 | 1281085 |
| π 2026-08-19 (duwi) | AnTuTu 11 | 774697 |
| π 2026-08-17 -> Paranoid _Antost | AnTuTu 11 | 930801 |
| π 2026-08-04 (nguyen van nguyen) | AnTuTu 11 | 990021 |
| π 2026-07-29 (Mimo) | AnTuTu 11 | 1117985 |
Specifications
Full list of technical specifications of Tensor G5 and Google TensorCPU
| Architecture | 1x 3.78 GHz β Cortex-X4 5x 3.05 GHz β Cortex-A725 2x 2.25 GHz β Cortex-A520 |
2x 2.8 GHz β Cortex-X1 2x 2.25 GHz β Cortex A76 4x 1.8 GHz β Cortex A55 |
| Cores | 8 | 8 |
| Frequency | 3780 MHz | 2800 MHz |
| Instruction set | ARMv9.2-A | ARMv8.2-A |
| L2 cache | - | 3 MB |
| L3 cache | 8 MB | 4 MB |
| Process | 3 nanometers | 5 nanometers |
| TDP (Sustained Power Limit) | - | 5.6 W |
| Manufacturing | TSMC | Samsung |
Graphics
| GPU name | PowerVR DXT-48-1536 | Mali-G78 MP20 |
| Architecture | - | Valhall 2nd gen |
| GPU frequency | 1100 MHz | 850 MHz |
| Pipelines | 6 | 20 |
| Shading units | 128 | 32 |
| Total shaders | 768 | 640 |
| FLOPS | 1689.6 Gigaflops | 1085.4 Gigaflops |
| Vulkan version | 1.4 | 1.3 |
| OpenCL version | 3.0 | 2.0 |
AI Accelerator
| Neural processor (NPU) | Yes | Tensor Processing Unit |
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 | 12 GB |
Multimedia (ISP)
| Storage type | UFS 3.1, UFS 4.0 | UFS 3.1 |
| Max display resolution | 3840 x 2400 | 3840 x 2160 |
| Max camera resolution | 1x 200MP | 1x 200MP, 2x 32MP |
| 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 - VP9 |
| Audio codecs | - AAC - AIFF - CAF - MP3 - MP4 - WAV |
- AAC - AIFF - CAF - MP3 - MP4 - WAV |
Connectivity
| Modem | Exynos 5400c | Exynos 5300b |
| 4G support | - | LTE Cat. 24 |
| 5G support | Yes | Yes |
| Download speed | - | Up to 7350 Mbps |
| Upload speed | - | Up to 3670 Mbps |
| Wi-Fi | 7 | 6 |
| Bluetooth | 6.0 | 5.2 |
| Navigation | GPS, GLONASS, Beidou, Galileo, QZSS | GPS, GLONASS, Beidou, Galileo, QZSS |
Info
| Announced | August 2025 | October 2021 |
| Class | Flagship | Flagship |
| Model number | GS501 | GS101 (S5P9845) |
| Official page | - | Google Tensor official site |
| Successor | - | - Google Tensor G5 |
Cast your vote
Total votes: 48
