Quick/Comparisons
21 | Gaming | 22 |
32 | Workstation | 28 |
33 | AI/ML |
Compare with something else?
Compare Apple M4 Pro with another →16 |
94 | Energy Efficiency | 20 |
37 | Quick Comparison Final Score | 23 |
| Custom | Graphics Processor | GP104 |
| 2048 | Cores | 2432 |
| 128 / 64 | TMUs / ROPs | 152 / 64 |
| 2355.52 | Blender GPU | 629.59 |
| — | Forza Horizon 5 | 55 |
| — | The Witcher 3 | 77 |
| — | Counter-Strike 2 | 108 |
| — | Far Cry 6 | 70 |
| — | Hogwarts Legacy | 42 |
| — | Call of Duty | 53 |
| — | Ghost of Tsushima | 39 |
| — | Cyberpunk 2077 | 45 |
| — | Tomb Raider | 80 |
| — | Average FPS | 63 |
| — | 1080p High | 79 |
| — | 1080p Ultra | 63 |
| — | 1440p Ultra | 45 |
| — | 4K Ultra | 24 |
| — | Margin of Error | Low |
| 61008 | GB6 Compute Score | 57735 |
| 106 img/sec | Background Blur | 149.2 img/sec |
| 73.1 img/sec | Face Detection | 67.6 img/sec |
| 2.59 Gpixels/sec | Horizon Detection | 2.36 Gpixels/sec |
| 3.8 Gpixels/sec | Edge Detection | 3.57 Gpixels/sec |
| 2.4 Gpixels/sec | Gaussian Blur | 4.32 Gpixels/sec |
| 0.73 Gpixels/sec | Feature Matching | 0.27 Gpixels/sec |
| 198.6 Gpixels/sec | Stereo Matching | 209.6 Gpixels/sec |
| 8166.3 FPS | Particle Physics | 6584.9 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 14658 |
| — | G2D Mark | 876 |
| — | DirectX 11 | 107 FPS |
| — | DirectX 12 | 52 FPS |
| — | GPU Compute | 7162 Ops/s |
| 16605 | GB6 ML Single Precision | 8949 |
| 18809 | GB6 ML Half Precision | 8769 |
| 17506 | GB6 ML Quantized | 6546 |
| 7548 | Image Classification (SP) | 3473 |
| 15108 | Image Segmentation (HP) | 5149 |
| 20789 | Image Super Resolution (Q) | 9722 |
| 28665 | Face Detection (HP) | 9306 |
| 57124 | Pose Estimation (Q) | 34870 |
| 2888 | Text Classification (SP) | 2139 |
| 6086 | Machine Translation (HP) | 2424 |
| 6804 | Object Detection (SP) | 4566 |
| 32732 | Depth Estimation (Q) | 14316 |
| 123465 | Style Transfer (SP) | 88984 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 101 GPixel/s | Pixel Fill Rate | 108 GPixel/s |
| 202 GTexel/s | Texture Fill Rate | 256 GTexel/s |
| 6.5 TFLOPS | FLOPS (FP32) | 8.2 TFLOPS |
| — | Memory Size | 8 GB |
| 8533 MHz | Memory Clock | 2002 MHz |
| 15000 Mbps | Effective Memory Speed | 8000 Mbps |
| 192-bit | Bus | 256-bit |
| No | ECC | No |
| 273 GB/s | Memory Bandwidth | 256.3 GB/s |
| System Shared | Memory Type | GDDR5 |
| Custom | Interface | PCIe 3.0 x16 |
| 26 W | TGP | 180 W |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 16 nm |
| — | Die Size | 314 mm² |
| — | Transistor Count | 7 billion |
| — | Transistor Density | 22.29 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.3 |
| — | OpenGL | 4.6 |
| — | OpenCL | 3 |
| — | CUDA | 6.1 |
| Yes | Ray Tracing | No |
| No | DLSS | No |
| — | DisplayPort | 1.4a |
| Apple | Vendor | Nvidia |
| Integrated | Build | Discrete |
| October 29, 2024 | Released | November 2, 2017 |
| Laptop | Case | Desktop |
| Professional | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Apple M GPU | Architecture | Pascal |
| Custom | GPU Codename | GP104 |
| -Apple M4 Pro (12-Core)or above | Recommended CPU | -Intel Core i7 7700Kor above |
| 6944 | Steel Nomad Lite Score | 6671 |
| — | Time Spy | 6843 |
| 28721 | Solar Bay | 18267 |
| — | Port Royal | 1552 |
| — | Fire Strike | 20115 |
| 16466 | Wild Life Extreme | 13248 |
| — | Night Raid | 78890 |
| 500 MHz | Base Clock | 1607 MHz |
| 1578 MHz | Boost Clock | 1683 MHz |
| 2048 | Shading Units | 2432 |
| 128 | Texture Mapping Units (TMUs) | 152 |
| 64 | Render Output Units (ROPs) | 64 |
| 256 | Compute Units (Pipelines) | 19 |
| — | Tensor Cores | No |
| — | Ray-tracing Cores | No |
| — | L1 Cache | 48KB per cluster |
| — | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |