Quick/Comparisons
14 | Gaming | 35 |
24 | Workstation | 43 |
| — | AI/ML |
Compare with something else?
Compare Apple M4 GPU with another →97 | Energy Efficiency | 41 |
32 | Quick Comparison Final Score | 37 |
| Custom | Graphics Processor | DG2-512 |
| 1024 | Cores | 4096 |
| 64 / 32 | TMUs / ROPs | 256 / 128 |
| 1060.26 | Blender GPU | 2016.55 |
| — | Forza Horizon 5 | 77 |
| — | The Witcher 3 | 124 |
| — | Counter-Strike 2 | 118 |
| — | Far Cry 6 | 97 |
| — | Hogwarts Legacy | 59 |
| — | Call of Duty | 79 |
| — | Ghost of Tsushima | 57 |
| — | Cyberpunk 2077 | 80 |
| — | Tomb Raider | 106 |
| — | Average FPS | 89 |
| — | 1080p High | 106 |
| — | 1080p Ultra | 89 |
| — | 1440p Ultra | 65 |
| — | 4K Ultra | 37 |
| — | Margin of Error | Low |
| 30791 | GB6 Compute Score | 113935 |
| 57.4 img/sec | Background Blur | 175 img/sec |
| 41.3 img/sec | Face Detection | 98 img/sec |
| 1.21 Gpixels/sec | Horizon Detection | 3.63 Gpixels/sec |
| 1.74 Gpixels/sec | Edge Detection | 5.51 Gpixels/sec |
| 1.12 Gpixels/sec | Gaussian Blur | 8.51 Gpixels/sec |
| 0.43 Gpixels/sec | Feature Matching | 0.89 Gpixels/sec |
| 94.9 Gpixels/sec | Stereo Matching | 492.6 Gpixels/sec |
| 3902.5 FPS | Particle Physics | 24819.2 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 13358 |
| — | G2D Mark | 761 |
| — | DirectX 11 | 78 FPS |
| — | DirectX 12 | 75 FPS |
| — | GPU Compute | 5850 Ops/s |
| — | GB6 ML Single Precision | 13344 |
| — | GB6 ML Half Precision | 14119 |
| — | GB6 ML Quantized | 10082 |
| — | Image Classification (SP) | 5750 |
| — | Image Segmentation (HP) | 16175 |
| — | Image Super Resolution (Q) | 12997 |
| — | Face Detection (HP) | 21409 |
| — | Pose Estimation (Q) | 67820 |
| — | Text Classification (SP) | 849 |
| — | Machine Translation (HP) | 2627 |
| — | Object Detection (SP) | 7465 |
| — | Depth Estimation (Q) | 24186 |
| — | Style Transfer (SP) | 183238 |
| — | Framework | ONNX |
| — | Backend | DirectML |
| 47 GPixel/s | Pixel Fill Rate | 307 GPixel/s |
| 94 GTexel/s | Texture Fill Rate | 614 GTexel/s |
| 3 TFLOPS | FLOPS (FP32) | 19.7 TFLOPS |
| System Shared | Memory Type | GDDR6 |
| — | Memory Size | 16 GB |
| 7500 MHz | Memory Clock | 2000 MHz |
| 15000 Mbps | Effective Memory Speed | 16000 Mbps |
| 128-bit | Bus | 256-bit |
| No | ECC | No |
| 120 GB/s | Memory Bandwidth | 512 GB/s |
| Custom | Interface | PCIe 4.0 x16 |
| 15 W | TGP | 225 W |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 6 nm |
| — | Die Size | 406 mm² |
| — | Transistor Count | 21 billion |
| — | Transistor Density | 51.72 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.3 |
| — | OpenGL | 4.6 |
| — | OpenCL | 3 |
| — | CUDA | — |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| — | DisplayPort | — |
| Apple | Vendor | Intel |
| Integrated | Build | Discrete |
| October 29, 2024 | Released | October 12, 2022 |
| Laptop | Case | Desktop |
| Professional | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Apple M GPU | Architecture | Generation 12.7 |
| Custom | GPU Codename | DG2-512 |
| -Apple M4 (8-Core)or above | Recommended CPU | -Intel Core i5 12500or above |
| 3450 | Steel Nomad Lite Score | 12648 |
| — | Time Spy | 13554 |
| 13908 | Solar Bay | 51114 |
| — | Port Royal | 7231 |
| — | Fire Strike | 34163 |
| 7928 | Wild Life Extreme | 26266 |
| — | Night Raid | 122588 |
| 500 MHz | Base Clock | 2100 MHz |
| 1470 MHz | Boost Clock | 2400 MHz |
| 1024 | Shading Units | 4096 |
| 64 | Texture Mapping Units (TMUs) | 256 |
| 32 | Render Output Units (ROPs) | 128 |
| 128 | Compute Units (Pipelines) | — |
| — | Tensor Cores | 512 |
| — | Ray-tracing Cores | 32 |
| — | L1 Cache | — |
| — | L2 Cache | 16MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |