Quick/Comparisons
14 | Gaming | 66 |
24 | Workstation | 73 |
| — | AI/ML |
Compare with something else?
Compare Apple M4 GPU with another →97 | Energy Efficiency | 37 |
32 | Quick Comparison Final Score | 66 |
| Custom | Graphics Processor | GA102 |
| 1024 | Cores | 10752 |
| 64 / 32 | TMUs / ROPs | 336 / 112 |
| 1060.26 | Blender GPU | 6057.83 |
| — | Forza Horizon 5 | 141 |
| — | The Witcher 3 | 182 |
| — | Counter-Strike 2 | 210 |
| — | Far Cry 6 | 147 |
| — | Hogwarts Legacy | 112 |
| — | Call of Duty | 173 |
| — | Ghost of Tsushima | 108 |
| — | Cyberpunk 2077 | 143 |
| — | Tomb Raider | 243 |
| — | Average FPS | 162 |
| — | 1080p High | 193 |
| — | 1080p Ultra | 162 |
| — | 1440p Ultra | 128 |
| — | 4K Ultra | 75 |
| — | Margin of Error | Low |
| 30791 | GB6 Compute Score | 232083 |
| 57.4 img/sec | Background Blur | 362.7 img/sec |
| 41.3 img/sec | Face Detection | 227.9 img/sec |
| 1.21 Gpixels/sec | Horizon Detection | 11.3 Gpixels/sec |
| 1.74 Gpixels/sec | Edge Detection | 15.9 Gpixels/sec |
| 1.12 Gpixels/sec | Gaussian Blur | 12.1 Gpixels/sec |
| 0.43 Gpixels/sec | Feature Matching | 1.56 Gpixels/sec |
| 94.9 Gpixels/sec | Stereo Matching | 1080 Gpixels/sec |
| 3902.5 FPS | Particle Physics | 31024.6 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 29288 |
| — | G2D Mark | 1226 |
| — | DirectX 11 | 238 FPS |
| — | DirectX 12 | 122 FPS |
| — | GPU Compute | 16569 Ops/s |
| — | GB6 ML Single Precision | 26701 |
| — | GB6 ML Half Precision | 38573 |
| — | GB6 ML Quantized | 15995 |
| — | Image Classification (SP) | 10246 |
| — | Image Segmentation (HP) | 22830 |
| — | Image Super Resolution (Q) | 23532 |
| — | Face Detection (HP) | 48657 |
| — | Pose Estimation (Q) | 147241 |
| — | Text Classification (SP) | 2917 |
| — | Machine Translation (HP) | 4427 |
| — | Object Detection (SP) | 13323 |
| — | Depth Estimation (Q) | 40055 |
| — | Style Transfer (SP) | 298683 |
| — | Framework | ONNX |
| — | Backend | DirectML |
| 47 GPixel/s | Pixel Fill Rate | 208 GPixel/s |
| 94 GTexel/s | Texture Fill Rate | 625 GTexel/s |
| 3 TFLOPS | FLOPS (FP32) | 40 TFLOPS |
| System Shared | Memory Type | GDDR6X |
| — | Memory Size | 24 GB |
| 7500 MHz | Memory Clock | 1313 MHz |
| 15000 Mbps | Effective Memory Speed | 21000 Mbps |
| 128-bit | Bus | 384-bit |
| No | ECC | No |
| 120 GB/s | Memory Bandwidth | 1008 GB/s |
| Custom | Interface | PCIe 4.0 x16 |
| 15 W | TGP | 450 W |
| TSMC | Manufacturing | Samsung |
| 3 nm | Fabrication Process | 8 nm |
| — | Die Size | 628 mm² |
| — | Transistor Count | 28 billion |
| — | Transistor Density | 44.59 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.3 |
| — | OpenGL | 4.6 |
| — | OpenCL | 3 |
| — | CUDA | 8.6 |
| Yes | Ray Tracing | Yes |
| No | DLSS | DLSS 2 |
| — | DisplayPort | 1.4a |
| Apple | Vendor | Nvidia |
| Integrated | Build | Discrete |
| October 29, 2024 | Released | March 29, 2022 |
| Laptop | Case | Desktop |
| Professional | Purpose | Gaming |
| Mid-range | Segment | High-end |
| Apple M GPU | Architecture | Ampere |
| Custom | GPU Codename | GA102 |
| -Apple M4 (8-Core)or above | Recommended CPU | -Intel Core i9 12900Kor above |
| 3450 | Steel Nomad Lite Score | 24378 |
| — | Time Spy | 21854 |
| 13908 | Solar Bay | 109167 |
| — | Port Royal | 14888 |
| — | Fire Strike | 51943 |
| 7928 | Wild Life Extreme | 48864 |
| — | Night Raid | 156500 |
| 500 MHz | Base Clock | 1560 MHz |
| 1470 MHz | Boost Clock | 1860 MHz |
| 1024 | Shading Units | 10752 |
| 64 | Texture Mapping Units (TMUs) | 336 |
| 32 | Render Output Units (ROPs) | 112 |
| 128 | Compute Units (Pipelines) | 84 |
| — | Tensor Cores | 336 |
| — | Ray-tracing Cores | 84 |
| — | L1 Cache | 128KB per cluster |
| — | L2 Cache | 6MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |