Quick/Comparisons
50 | Gaming | 6 |
59 | Workstation | 9 |
41 | AI/ML |
Compare with something else?
Compare Apple M3 Ultra with another →9 |
61 | Energy Efficiency | 86 |
52 | Quick Comparison Final Score | 20 |
| Custom | Graphics Processor | Krackan Point |
| 10240 | Cores | 256 |
| 640 / 320 | TMUs / ROPs | 16 / 8 |
| 7360.18 | Blender GPU | 122.02 |
| — | Forza Horizon 5 | 13 |
| — | The Witcher 3 | 12 |
| — | Counter-Strike 2 | 18 |
| — | Far Cry 6 | 16 |
| — | Hogwarts Legacy | - |
| — | Call of Duty | 19 |
| — | Ghost of Tsushima | 9 |
| — | Cyberpunk 2077 | 9 |
| — | Tomb Raider | 17 |
| — | Average FPS | 13 |
| — | 1080p High | 18 |
| — | 1080p Ultra | 13 |
| — | 1440p Ultra | 10 |
| — | 4K Ultra | - |
| — | Margin of Error | High |
| 147719 | GB6 Compute Score | 14025 |
| 204.5 img/sec | Background Blur | 31.8 img/sec |
| 126 img/sec | Face Detection | 16.8 img/sec |
| 6.53 Gpixels/sec | Horizon Detection | 0.46 Gpixels/sec |
| 10.7 Gpixels/sec | Edge Detection | 0.58 Gpixels/sec |
| 9.5 Gpixels/sec | Gaussian Blur | 0.42 Gpixels/sec |
| 1.05 Gpixels/sec | Feature Matching | 0.18 Gpixels/sec |
| 560.3 Gpixels/sec | Stereo Matching | 52.5 Gpixels/sec |
| 25267.8 FPS | Particle Physics | 2048.6 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | — |
| — | G2D Mark | — |
| — | DirectX 11 | — |
| — | DirectX 12 | — |
| — | GPU Compute | — |
| 21356 | GB6 ML Single Precision | 4884 |
| 23530 | GB6 ML Half Precision | 6067 |
| 21706 | GB6 ML Quantized | 3946 |
| 8094 | Image Classification (SP) | 1988 |
| 24595 | Image Segmentation (HP) | 2184 |
| 25190 | Image Super Resolution (Q) | 4788 |
| 38337 | Face Detection (HP) | 6857 |
| 111021 | Pose Estimation (Q) | 17396 |
| 2786 | Text Classification (SP) | 1628 |
| 4339 | Machine Translation (HP) | 2053 |
| 7668 | Object Detection (SP) | 1825 |
| 37488 | Depth Estimation (Q) | 8471 |
| 244483 | Style Transfer (SP) | 31032 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 442 GPixel/s | Pixel Fill Rate | 23 GPixel/s |
| 883 GTexel/s | Texture Fill Rate | 46 GTexel/s |
| 28.3 TFLOPS | FLOPS (FP32) | 1.5 TFLOPS |
| System Shared | Memory Type | System Shared |
| — | Memory Size | — |
| 6400 MHz | Memory Clock | — |
| — | Effective Memory Speed | — |
| 1024-bit | Bus | — |
| No | ECC | — |
| 819.3 GB/s | Memory Bandwidth | — |
| Custom | Interface | PCIe 4.0 x16 |
| 140 W | TGP | 15-54 W (configurable) |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 4 nm |
| — | Die Size | — |
| 184 billion | Transistor Count | — |
| — | Transistor Density | — |
| — | DirectX | 12 |
| — | Vulkan | 1.4 |
| — | OpenGL | 4.6 |
| — | OpenCL | 2.1 |
| — | CUDA | No |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| — | DisplayPort | 2.1 |
| Apple | Vendor | Amd |
| Integrated | Build | Integrated |
| March 5, 2025 | Released | January 7, 2025 |
| Desktop | Case | Laptop |
| Professional | Purpose | Generic |
| High-end | Segment | Entry-level |
| Apple M GPU | Architecture | RDNA 3.5 |
| Custom | GPU Codename | Krackan Point |
| -Apple M3 Ultraor above | Recommended CPU | -AMD Ryzen AI 5 340or above |
| 19491 | Steel Nomad Lite Score | 1293 |
| — | Time Spy | 1700 |
| 73690 | Solar Bay | 7155 |
| — | Port Royal | 577 |
| — | Fire Strike | 5317 |
| 44759 | Wild Life Extreme | — |
| — | Night Raid | 22727 |
| — | Base Clock | 400 MHz |
| 1380 MHz | Boost Clock | 2900 MHz |
| 10240 | Shading Units | 256 |
| 640 | Texture Mapping Units (TMUs) | 16 |
| 320 | Render Output Units (ROPs) | 8 |
| 1280 | Compute Units (Pipelines) | 4 |
| — | Tensor Cores | No |
| — | Ray-tracing Cores | 4 |
| — | L1 Cache | 512KB per cluster |
| — | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |