Apple A19 Pro vs Google Tensor
Across 12 measured categories, the Apple A19 Pro comes out ahead.
I put the Apple A19 Pro and the Google Tensor side by side to see how they really compare. On graphics, the Apple A19 Pro runs the Apple A19 pro (6-core GPU) while the Google Tensor runs the Mali-G78 MP20. Below I go through the CPU, GPU, memory and connectivity, then the benchmark scores, so you can work out which one is the better fit for you.
Where each chip wins
Apple A19 Pro
10 advantages
-
103% higher AnTuTu score
20.07 Lakhs vs 9.87 Lakhs
-
204% faster Geekbench multi-core score
10,009 vs 3,294
-
180% faster Geekbench single-core score
3,879 vs 1,384
-
1,460 MHz higher CPU clock
4260 MHz vs 2800 MHz
-
Newer CPU architecture
ARMv9.2-A vs ARMv8.2-A
-
Larger L3 cache
32 MB vs 4 MB
-
2 nm smaller semiconductor size
3 nm vs 5 nm
-
Higher GPU FLOPS
2488.3 GFLOPS vs 1085 GFLOPS
-
Newer memory type
LPDDR5X vs LPDDR5
-
1,600 MHz higher RAM speed
4800 MHz vs 3200 MHz
Google Tensor
0 advantages
- No standout advantages for this chip.
Benchmark charts
AnTuTu v11 Total
Geekbench Single-Core
Geekbench Multi-Core
Specifications Comparison
Here is how the Apple A19 Pro and the Google Tensor compare across every spec. In each row, the stronger value is highlighted so the differences are easy to spot.
| Launch Date | September 2025 | October 2021 |
|---|---|---|
| Chipset Class | Flagship | Flagship |
| Manufacturer | Apple | |
| Model Number | T8150 | GS101 |
| Total Cores | 6 | 8 |
|---|---|---|
| Core Configuration | 2x 4.26 GHz performance cores 4x 2.60 GHz efficiency cores |
2x 2.8 GHz - Cortex X1 2x 2.25 GHz - Cortex A76 4x 1.8 GHz - Cortex A55 |
| Max Frequency | 4260 MHz | 2800 MHz |
| Instruction Set | ARMv9.2-A | ARMv8.2-A |
| Process Node | 3 nm | 5 nm |
| Manufacturing | TSMC | Samsung Foundry |
| L2 Cache | 16 MB (performance) 4 MB (efficiency) |
3 MB |
| L3 Cache | 32 MB | 4 MB |
| TDP | - | 5.6 W |
| GPU Name | Apple A19 pro (6-core GPU) | Mali-G78 MP20 |
|---|---|---|
| GPU Frequency | 1620 MHz | 850 MHz |
| Shading Units | 128 | 640 |
| FLOPS | 2488.3 GFLOPS | 1085 GFLOPS |
| Ray Tracing Support | Yes | No |
| NPU | Neural Engine | Edge TPU |
| API Support | Metal 3, Vulkan 1.3, OpenGL ES 3.2, OpenCL 3.0 | Vulkan 1.3, Open CL 2.0 |
| RAM Type | LPDDR5X | LPDDR5 |
|---|---|---|
| RAM Frequency | 4800 MHz | 3200 MHz |
| Memory Bus | 4x 16 Bit | 4x16-bit |
| Max Bandwidth | 75.8 GB/s | 51.2 GB/s |
| Max RAM Size | 12 GB | 12 GB |
| Storage Type | NVMe | UFS 3.1 |
| Camera Support | 48MP | 200 MP |
|---|---|---|
| Video Capture | 4K at 120FPS | 4K at 60FPS, 1080p at 240FPS |
| Max Display Resolution | 4K | 4K |
| Video Playback | 4K at 120FPS | 4K at 60FPS |
| Video Codecs | H.264, H.265/HEVC, AV1, VP8, VP9 | H.264, H.265, VP9, AV1 |
| Audio Codecs | AAC, AAX, AAX+, AC-3, AIFF, CAF, E-AC-3, MP3, MP4, WAV | AAC, AIFF, CAF, MP3, MP4, WAV |
| Modem | Snapdragon (on Pro models), Apple C1X (on Air) | Exynos 5300b 5G |
|---|---|---|
| Network Support | 5G | 5G |
| Download Speed | 10 Gbps (Snapdragon) | 7.35 Gbps |
| Upload Speed | 3.5 Gbps (Snapdragon) | 3.67 Gbps |
| Wi-Fi | 7 | 6 |
| Bluetooth | 6.0 | 5.2 |
| Navigation | NAVIC, GPS, GLONASS, Beidou, Galileo, QZSS | GPS, GLONASS, BeiDou, Galileo, QZSS |
Benchmark results
| Total | 20,06,613 | 9,87,312 |
|---|---|---|
| CPU | 7,96,137 | 3,88,463 |
| GPU | 6,02,282 | 1,86,800 |
| Memory | 2,44,562 | 1,64,486 |
| UX | 3,61,022 | 2,47,563 |
| Single-Core | 3,879 | 1,384 |
|---|---|---|
| Multi-Core | 10,009 | 3,294 |
Overall Score
Each category scored out of 100.
| Category | Apple A19 Pro | Google Tensor |
|---|---|---|
| CPU Score |
|
|
| GPU Score |
|
|
| Memory Score |
|
|
| Total Score |
|
|
Related comparisons
- 1 Apple A19 Pro vs Snapdragon 7s Gen 3
- 2 Apple A19 Pro vs MediaTek Dimensity 9000
- 3 Apple A19 Pro vs Apple A15 Bionic
- 4 Apple A19 Pro vs Google Tensor G2
- 5 Apple A19 Pro vs Snapdragon 7 Gen 4
- 6 Exynos 2600 vs Google Tensor
- 7 Xiaomi Xring O1 vs Google Tensor
- 8 MediaTek Dimensity 9300 vs Google Tensor
- 9 Apple A18 Pro vs Google Tensor
- 10 Snapdragon 8 Elite vs Google Tensor
Discussion