Tensor G4 vs Tensor G2
We compared two 8-core processors: Google Tensor G4 (with Mali-G715 MP7 graphics) and Tensor G2 (Mali-G710 MP7). 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 G4
- Performs 2.2x better in floating-point computations
- Has a smaller size transistor (4 versus 5 nm)
- Supports 33% higher memory bandwidth (68.2 against 51.2 GB/s)
- Shows better (up to 28%) AnTuTu 11 score β 1519K vs 1186K
- 9% higher CPU clock speed (3100 vs 2850 MHz)
- Better instruction set architecture
- Higher GPU frequency (~11%)
Benchmarks
Performance tests in popular benchmarksAnTuTu 11
| CPU | 536742 | 384056 |
| GPU | 414930 | 278383 |
| Memory | 215572 | 220709 |
| UX | 352693 | 303255 |
| Total score | 1519937 | 1186373 |
| Samples | [25 tests] | [54 tests] |
GeekBench 6
Single-Core Score
Multi-Core Score
| Asset compression | 177.3 MB/sec | 149.9 MB/sec |
| HTML 5 Browser | 134 pages/sec | 99.9 pages/sec |
| PDF Renderer | 157 Mpixels/sec | 110.2 Mpixels/sec |
| Image detection | 122 images/sec | 74.2 images/sec |
| HDR | 128.3 Mpixels/sec | 105.5 Mpixels/sec |
| Background blur | 15.8 images/sec | 11.7 images/sec |
| Photo processing | 39.2 images/sec | 35.1 images/sec |
| Ray tracing | 4.65 Mpixels/sec | 4.26 Mpixels/sec |
| Samples | 11 tests | 26 tests |
Compute Score (GPU)
| Background blur | 13.4 images/sec | 6.68 images/sec |
| Face detection | 8.51 images/sec | 3.96 images/sec |
| Horizon detection | 0.32 Gpixels/sec | 0.22 Gpixels/sec |
| Edge detection | 0.3 Gpixels/sec | 0.35 Gpixels/sec |
| Gaussian blur | 0.36 Gpixels/sec | 0.2 Gpixels/sec |
| Feature matching | 0.11 Gpixels/sec | 0.06 Gpixels/sec |
| Stereo matching | 26.4 Gpixels/sec | 11.5 Gpixels/sec |
| Particle physics | 1273.7 FPS | 972.6 FPS |
| API | OpenCL | Vulkan |
| Samples | 9 tests | 17 tests |
3DMark
3DMark Steel Nomad Light
| Score | 1049 | 702 |
| Stability | 84% | 78% |
| Graphics test | 8 FPS | 5 FPS |
3DMark Wild Life Extreme
| Score | - | 1821 |
| Stability | - | 77% |
| Graphics test | - | 10 FPS |
PCMark 3.0
PCMark 3.0
| Web score | 10521 | 7923 |
| Video editing score | 7569 | 6434 |
| Data manipulation score | 14718 | 11061 |
| Photo editing score | 21074 | 18842 |
| Writing score | 15880 | 15053 |
| Phones tested | 4 models | 5 models |
Recent User Tests
Google Tensor G4
| Date | Benchmark | Result |
|---|---|---|
| π 2026-09-18 -> Oleg 16 cm | AnTuTu 11 | 1423579 |
| π 2026-08-27 (feraroof43) | AnTuTu 11 | 1626655 |
| π 2026-08-21 (Leif) | AnTuTu 11 | 1590049 |
| π 2026-08-21 (Kazin) | AnTuTu 11 | 1723964 |
| π 2026-07-14 (123) | Geekbench 6 GPU | 8744 |
Google Tensor G2
| Date | Benchmark | Result |
|---|---|---|
| π 2026-09-28 -> putkin_prosrochka59 | AnTuTu 12 | 1477460 |
| π 2026-09-09 -> Kimash333 | AnTuTu 11 | 1253847 |
| π 2026-09-08 (Kimash333) | AnTuTu 11 | 1253847 |
| π 2026-09-02 -> nanamond | AnTuTu 11 | 1196742 |
| π 2026-08-06 -> shu- | AnTuTu 11 | 1126520 |
Specifications
Full list of technical specifications of Tensor G4 and Tensor G2CPU
| Architecture | 1x 3.1 GHz β Cortex-X4 3x 2.6 GHz β Cortex-A720 4x 1.92 GHz β Cortex-A520 |
2x 2.85 GHz β Cortex-X1 2x 2.35 GHz β Cortex-A78 4x 1.8 GHz β Cortex-A55 |
| Cores | 8 | 8 |
| Frequency | 3100 MHz | 2850 MHz |
| Instruction set | ARMv9.2-A | ARMv8.4-A |
| L3 cache | - | 4 MB |
| Process | 4 nanometers | 5 nanometers |
| TDP (Sustained Power Limit) | - | 7 W |
| Manufacturing | Samsung | Samsung |
Graphics
| GPU name | Mali-G715 MP7 | Mali-G710 MP7 |
| Architecture | Valhall 4th gen | Valhall 3rd gen |
| GPU frequency | 940 MHz | 848 MHz |
| Pipelines | 7 | 7 |
| Shading units | 128 | 64 |
| Total shaders | 896 | 448 |
| FLOPS | 1684.4 Gigaflops | 759.8 Gigaflops |
| Vulkan version | 1.3 | 1.3 |
| OpenCL version | 2.0 | 2.0 |
AI Accelerator
| Neural processor (NPU) | Yes | Next-gen 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 | 16 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, 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 5300g |
| 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, NAVIC |
Info
| Announced | August 2024 | October 2022 |
| Class | Flagship | Flagship |
| Model number | GS401 (S5P9875) | GS201 (S5P9855) |
| Successor | - Google Tensor G5 | - Google Tensor G5 |
Cast your vote
Total votes: 201