Quick/Comparisons
50 | Gaming | 31 |
59 | Workstation | 37 |
41 | AI/ML |
Compare with something else?
Compare Apple M3 Ultra with another →29 |
61 | Energy Efficiency | 37 |
52 | Quick Comparison Final Score | 33 |
| Custom | Graphics Processor | GA106 |
| 10240 | Cores | 3584 |
| 640 / 320 | TMUs / ROPs | 112 / 48 |
| 7360.18 | Blender GPU | 2303.56 |
| — | Forza Horizon 5 | 71 |
| — | The Witcher 3 | 100 |
| — | Counter-Strike 2 | 133 |
| — | Far Cry 6 | 83 |
| — | Hogwarts Legacy | 56 |
| — | Call of Duty | 73 |
| — | Ghost of Tsushima | 49 |
| — | Cyberpunk 2077 | 62 |
| — | Tomb Raider | 97 |
| — | Average FPS | 80 |
| — | 1080p High | 100 |
| — | 1080p Ultra | 80 |
| — | 1440p Ultra | 59 |
| — | 4K Ultra | 32 |
| — | Margin of Error | Low |
| 147719 | GB6 Compute Score | 91603 |
| 204.5 img/sec | Background Blur | 149.6 img/sec |
| 126 img/sec | Face Detection | 91 img/sec |
| 6.53 Gpixels/sec | Horizon Detection | 4.07 Gpixels/sec |
| 10.7 Gpixels/sec | Edge Detection | 5.58 Gpixels/sec |
| 9.5 Gpixels/sec | Gaussian Blur | 3.92 Gpixels/sec |
| 1.05 Gpixels/sec | Feature Matching | 1.01 Gpixels/sec |
| 560.3 Gpixels/sec | Stereo Matching | 372.9 Gpixels/sec |
| 25267.8 FPS | Particle Physics | 12226.3 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 16744 |
| — | G2D Mark | 966 |
| — | DirectX 11 | 128 FPS |
| — | DirectX 12 | 69 FPS |
| — | GPU Compute | 8266 Ops/s |
| 21356 | GB6 ML Single Precision | 15123 |
| 23530 | GB6 ML Half Precision | 25942 |
| 21706 | GB6 ML Quantized | 10356 |
| 8094 | Image Classification (SP) | 6645 |
| 24595 | Image Segmentation (HP) | 15338 |
| 25190 | Image Super Resolution (Q) | 14283 |
| 38337 | Face Detection (HP) | 28862 |
| 111021 | Pose Estimation (Q) | 61313 |
| 2786 | Text Classification (SP) | 2463 |
| 4339 | Machine Translation (HP) | 3793 |
| 7668 | Object Detection (SP) | 7712 |
| 37488 | Depth Estimation (Q) | 25461 |
| 244483 | Style Transfer (SP) | 123233 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 442 GPixel/s | Pixel Fill Rate | 85 GPixel/s |
| 883 GTexel/s | Texture Fill Rate | 199 GTexel/s |
| 28.3 TFLOPS | FLOPS (FP32) | 12.7 TFLOPS |
| System Shared | Memory Type | GDDR6 |
| — | Memory Size | 12 GB |
| 6400 MHz | Memory Clock | 1875 MHz |
| — | Effective Memory Speed | 15000 Mbps |
| 1024-bit | Bus | 192-bit |
| No | ECC | No |
| 819.3 GB/s | Memory Bandwidth | 360 GB/s |
| Custom | Interface | PCIe 4.0 x16 |
| 140 W | TGP | 170 W |
| TSMC | Manufacturing | Samsung |
| 3 nm | Fabrication Process | 8 nm |
| — | Die Size | 276 mm² |
| 184 billion | Transistor Count | 12 billion |
| — | Transistor Density | 43.48 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 |
| March 5, 2025 | Released | January 12, 2021 |
| Desktop | Case | Desktop |
| Professional | Purpose | Gaming |
| High-end | Segment | Mid-range |
| Apple M GPU | Architecture | Ampere |
| Custom | GPU Codename | GA106 |
| -Apple M3 Ultraor above | Recommended CPU | -Intel Core i5 12600Kor above |
| 19491 | Steel Nomad Lite Score | 9002 |
| — | Time Spy | 8754 |
| 73690 | Solar Bay | 41139 |
| — | Port Royal | 5145 |
| — | Fire Strike | 22306 |
| 44759 | Wild Life Extreme | 17726 |
| — | Night Raid | 97592 |
| — | Base Clock | 1320 MHz |
| 1380 MHz | Boost Clock | 1777 MHz |
| 10240 | Shading Units | 3584 |
| 640 | Texture Mapping Units (TMUs) | 112 |
| 320 | Render Output Units (ROPs) | 48 |
| 1280 | Compute Units (Pipelines) | 28 |
| — | Tensor Cores | 112 |
| — | Ray-tracing Cores | 28 |
| — | L1 Cache | 128KB per cluster |
| — | L2 Cache | 3MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |