Quick/Comparisons
50 | Gaming | 9 |
59 | Workstation | 8 |
41 | AI/ML |
Compare with something else?
Compare Apple M3 Ultra with another →3 |
61 | Energy Efficiency | 73 |
52 | Quick Comparison Final Score | 19 |
| Custom | Graphics Processor | GP108 |
| 10240 | Cores | 256 |
| 640 / 320 | TMUs / ROPs | 16 / 16 |
| 7360.18 | Blender GPU | 41.33 |
| — | Forza Horizon 5 | — |
| — | The Witcher 3 | — |
| — | Counter-Strike 2 | — |
| — | Far Cry 6 | — |
| — | Hogwarts Legacy | — |
| — | Call of Duty | — |
| — | Ghost of Tsushima | — |
| — | Cyberpunk 2077 | — |
| — | Tomb Raider | — |
| — | Average FPS | — |
| — | 1080p High | — |
| — | 1080p Ultra | — |
| — | 1440p Ultra | — |
| — | 4K Ultra | — |
| — | Margin of Error | — |
| 147719 | GB6 Compute Score | 7785 |
| 204.5 img/sec | Background Blur | 19.4 img/sec |
| 126 img/sec | Face Detection | 9.09 img/sec |
| 6.53 Gpixels/sec | Horizon Detection | 0.29 Gpixels/sec |
| 10.7 Gpixels/sec | Edge Detection | 0.43 Gpixels/sec |
| 9.5 Gpixels/sec | Gaussian Blur | 0.44 Gpixels/sec |
| 1.05 Gpixels/sec | Feature Matching | 0.08 Gpixels/sec |
| 560.3 Gpixels/sec | Stereo Matching | 25 Gpixels/sec |
| 25267.8 FPS | Particle Physics | 759.2 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 1818 |
| — | G2D Mark | 213 |
| — | DirectX 11 | 15 FPS |
| — | DirectX 12 | 9 FPS |
| — | GPU Compute | 762 Ops/s |
| 21356 | GB6 ML Single Precision | 1408 |
| 23530 | GB6 ML Half Precision | 1384 |
| 21706 | GB6 ML Quantized | 1028 |
| 8094 | Image Classification (SP) | 277 |
| 24595 | Image Segmentation (HP) | 726 |
| 25190 | Image Super Resolution (Q) | 1787 |
| 38337 | Face Detection (HP) | 1076 |
| 111021 | Pose Estimation (Q) | 5990 |
| 2786 | Text Classification (SP) | 743 |
| 4339 | Machine Translation (HP) | 970 |
| 7668 | Object Detection (SP) | 379 |
| 37488 | Depth Estimation (Q) | 2059 |
| 244483 | Style Transfer (SP) | 11915 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 442 GPixel/s | Pixel Fill Rate | 24 GPixel/s |
| 883 GTexel/s | Texture Fill Rate | 24 GTexel/s |
| 28.3 TFLOPS | FLOPS (FP32) | 0.8 TFLOPS |
| System Shared | Memory Type | GDDR5 |
| — | Memory Size | 2 GB |
| 6400 MHz | Memory Clock | 1502 MHz |
| — | Effective Memory Speed | 6000 Mbps |
| 1024-bit | Bus | 64-bit |
| No | ECC | — |
| 819.3 GB/s | Memory Bandwidth | 48 GB/s |
| Custom | Interface | PCIe 3.0 x4 |
| 140 W | TGP | 10 W |
| TSMC | Manufacturing | Samsung |
| 3 nm | Fabrication Process | 14 nm |
| — | Die Size | 74 mm² |
| 184 billion | Transistor Count | 1.8 billion |
| — | Transistor Density | 24.32 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.2 |
| — | OpenGL | 4.6 |
| — | OpenCL | 1.2 |
| — | CUDA | 6.1 |
| Yes | Ray Tracing | No |
| No | DLSS | No |
| — | DisplayPort | 1.4 |
| Apple | Vendor | Nvidia |
| Integrated | Build | Discrete |
| March 5, 2025 | Released | February 21, 2019 |
| Desktop | Case | Laptop |
| Professional | Purpose | Gaming |
| High-end | Segment | Entry-level |
| Apple M GPU | Architecture | Pascal |
| Custom | GPU Codename | GP108 |
| -Apple M3 Ultraor above | Recommended CPU | -Intel Core i7 1065G7or above |
| 19491 | Steel Nomad Lite Score | — |
| — | Time Spy | — |
| 73690 | Solar Bay | — |
| — | Port Royal | — |
| — | Fire Strike | — |
| 44759 | Wild Life Extreme | — |
| — | Night Raid | — |
| — | Base Clock | 1519 MHz |
| 1380 MHz | Boost Clock | 1531 MHz |
| 10240 | Shading Units | 256 |
| 640 | Texture Mapping Units (TMUs) | 16 |
| 320 | Render Output Units (ROPs) | 16 |
| 1280 | Compute Units (Pipelines) | 2 |
| — | Tensor Cores | No |
| — | Ray-tracing Cores | No |
| — | L1 Cache | 48KB per cluster |
| — | L2 Cache | 512KB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |