Quick/Comparisons
50 | Gaming | 27 |
59 | Workstation | 36 |
41 | AI/ML |
Compare with something else?
Compare Apple M3 Ultra with another →23 |
61 | Energy Efficiency | 44 |
52 | Quick Comparison Final Score | 30 |
| Custom | Graphics Processor | DG2-512 |
| 10240 | Cores | 3072 |
| 640 / 320 | TMUs / ROPs | 192 / 96 |
| 7360.18 | Blender GPU | 1860.75 |
| — | Forza Horizon 5 | 59 |
| — | The Witcher 3 | 97 |
| — | Counter-Strike 2 | 99 |
| — | Far Cry 6 | 83 |
| — | Hogwarts Legacy | 44 |
| — | Call of Duty | 56 |
| — | Ghost of Tsushima | 48 |
| — | Cyberpunk 2077 | 61 |
| — | Tomb Raider | 83 |
| — | Average FPS | 70 |
| — | 1080p High | 88 |
| — | 1080p Ultra | 70 |
| — | 1440p Ultra | 51 |
| — | 4K Ultra | 27 |
| — | Margin of Error | Low |
| 147719 | GB6 Compute Score | 89982 |
| 204.5 img/sec | Background Blur | 135.5 img/sec |
| 126 img/sec | Face Detection | 76.3 img/sec |
| 6.53 Gpixels/sec | Horizon Detection | 2.81 Gpixels/sec |
| 10.7 Gpixels/sec | Edge Detection | 4.37 Gpixels/sec |
| 9.5 Gpixels/sec | Gaussian Blur | 6.43 Gpixels/sec |
| 1.05 Gpixels/sec | Feature Matching | 0.8 Gpixels/sec |
| 560.3 Gpixels/sec | Stereo Matching | 417.1 Gpixels/sec |
| 25267.8 FPS | Particle Physics | 17700.8 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 12072 |
| — | G2D Mark | 743 |
| — | DirectX 11 | 72 FPS |
| — | DirectX 12 | 65 FPS |
| — | GPU Compute | 4913 Ops/s |
| 21356 | GB6 ML Single Precision | 9364 |
| 23530 | GB6 ML Half Precision | 9631 |
| 21706 | GB6 ML Quantized | 7378 |
| 8094 | Image Classification (SP) | 4014 |
| 24595 | Image Segmentation (HP) | 7656 |
| 25190 | Image Super Resolution (Q) | 10138 |
| 38337 | Face Detection (HP) | 12636 |
| 111021 | Pose Estimation (Q) | 50065 |
| 2786 | Text Classification (SP) | 672 |
| 4339 | Machine Translation (HP) | 1569 |
| 7668 | Object Detection (SP) | 5323 |
| 37488 | Depth Estimation (Q) | 20334 |
| 244483 | Style Transfer (SP) | 132558 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 442 GPixel/s | Pixel Fill Rate | 192 GPixel/s |
| 883 GTexel/s | Texture Fill Rate | 384 GTexel/s |
| 28.3 TFLOPS | FLOPS (FP32) | 12.3 TFLOPS |
| System Shared | Memory Type | GDDR6 |
| — | Memory Size | 8 GB |
| 6400 MHz | Memory Clock | 2000 MHz |
| — | Effective Memory Speed | 16000 Mbps |
| 1024-bit | Bus | 256-bit |
| No | ECC | No |
| 819.3 GB/s | Memory Bandwidth | 512 GB/s |
| Custom | Interface | PCIe 4.0 x16 |
| 140 W | TGP | 175 W |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 6 nm |
| — | Die Size | 406 mm² |
| 184 billion | 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 | 2 |
| Apple | Vendor | Intel |
| Integrated | Build | Discrete |
| March 5, 2025 | Released | October 10, 2023 |
| Desktop | Case | Desktop |
| Professional | Purpose | Gaming |
| High-end | Segment | Mid-range |
| Apple M GPU | Architecture | Generation 12.7 |
| Custom | GPU Codename | DG2-512 |
| -Apple M3 Ultraor above | Recommended CPU | -Intel Core i5 13400or above |
| 19491 | Steel Nomad Lite Score | 9225 |
| — | Time Spy | 10820 |
| 73690 | Solar Bay | 40676 |
| — | Port Royal | 5636 |
| — | Fire Strike | 25198 |
| 44759 | Wild Life Extreme | 21117 |
| — | Night Raid | 104083 |
| — | Base Clock | 1700 MHz |
| 1380 MHz | Boost Clock | 2000 MHz |
| 10240 | Shading Units | 3072 |
| 640 | Texture Mapping Units (TMUs) | 192 |
| 320 | Render Output Units (ROPs) | 96 |
| 1280 | Compute Units (Pipelines) | — |
| — | Tensor Cores | 384 |
| — | Ray-tracing Cores | 24 |
| — | L1 Cache | — |
| — | L2 Cache | 8MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |