Quick/Comparisons
14 | Gaming | 11 |
24 | Workstation | 17 |
| — | AI/ML |
Compare with something else?
Compare Apple M4 GPU with another →97 | Energy Efficiency | 78 |
32 | Quick Comparison Final Score | 24 |
| Custom | Graphics Processor | ACM-G11 |
| 1024 | Cores | 1024 |
| 64 / 32 | TMUs / ROPs | 64 / 32 |
| 1060.26 | Blender GPU | 377 |
| — | Forza Horizon 5 | 29 |
| — | The Witcher 3 | 29 |
| — | Counter-Strike 2 | 28 |
| — | Far Cry 6 | 31 |
| — | Hogwarts Legacy | 12 |
| — | Call of Duty | 28 |
| — | Ghost of Tsushima | 17 |
| — | Cyberpunk 2077 | 16 |
| — | Tomb Raider | 30 |
| — | Average FPS | 24 |
| — | 1080p High | 32 |
| — | 1080p Ultra | 24 |
| — | 1440p Ultra | 17 |
| — | 4K Ultra | 9 |
| — | Margin of Error | Low |
| 30791 | GB6 Compute Score | 31544 |
| 57.4 img/sec | Background Blur | 70.6 img/sec |
| 41.3 img/sec | Face Detection | 32.2 img/sec |
| 1.21 Gpixels/sec | Horizon Detection | 0.47 Gpixels/sec |
| 1.74 Gpixels/sec | Edge Detection | 1.37 Gpixels/sec |
| 1.12 Gpixels/sec | Gaussian Blur | 2.15 Gpixels/sec |
| 0.43 Gpixels/sec | Feature Matching | 0.34 Gpixels/sec |
| 94.9 Gpixels/sec | Stereo Matching | 138.8 Gpixels/sec |
| 3902.5 FPS | Particle Physics | 7194.7 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 5115 |
| — | G2D Mark | 724 |
| — | DirectX 11 | 29 FPS |
| — | DirectX 12 | 24 FPS |
| — | GPU Compute | 3014 Ops/s |
| — | GB6 ML Single Precision | 5065 |
| — | GB6 ML Half Precision | 5437 |
| — | GB6 ML Quantized | 3699 |
| — | Image Classification (SP) | 2580 |
| — | Image Segmentation (HP) | 4626 |
| — | Image Super Resolution (Q) | 4983 |
| — | Face Detection (HP) | 5258 |
| — | Pose Estimation (Q) | 16072 |
| — | Text Classification (SP) | 669 |
| — | Machine Translation (HP) | 1373 |
| — | Object Detection (SP) | 3260 |
| — | Depth Estimation (Q) | 9652 |
| — | Style Transfer (SP) | 47484 |
| — | Framework | ONNX |
| — | Backend | DirectML |
| 47 GPixel/s | Pixel Fill Rate | 50-66 GPixel/s |
| 94 GTexel/s | Texture Fill Rate | 99-131 GTexel/s |
| 3 TFLOPS | FLOPS (FP32) | 3.2-4.2 TFLOPS |
| System Shared | Memory Type | GDDR6 |
| — | Memory Size | 4 GB |
| 7500 MHz | Memory Clock | 1500 MHz |
| 15000 Mbps | Effective Memory Speed | 12000 Mbps |
| 128-bit | Bus | 64-bit |
| No | ECC | No |
| 120 GB/s | Memory Bandwidth | 96 GB/s |
| Custom | Interface | PCIe 4.0 x8 |
| 15 W | TGP | 35-50 W (configurable) |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 6 nm |
| — | Die Size | 157 mm² |
| — | Transistor Count | 7.2 billion |
| — | Transistor Density | 45.86 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 |
| October 29, 2024 | Released | March 30, 2022 |
| Laptop | Case | Laptop |
| Professional | Purpose | Generic |
| Mid-range | Segment | Entry-level |
| Apple M GPU | Architecture | Intel Xe-HPG |
| Custom | GPU Codename | ACM-G11 |
| -Apple M4 (8-Core)or above | Recommended CPU | -Intel Core i5 1240Por above |
| 3450 | Steel Nomad Lite Score | 2994 |
| — | Time Spy | 3438 |
| 13908 | Solar Bay | 11630 |
| — | Port Royal | 386 |
| — | Fire Strike | 7938 |
| 7928 | Wild Life Extreme | 5456 |
| — | Night Raid | 36312 |
| 500 MHz | Base Clock | 1150-1550 MHz (configurable) |
| 1470 MHz | Boost Clock | 1550-2050 MHz (configurable) |
| 1024 | Shading Units | 1024 |
| 64 | Texture Mapping Units (TMUs) | 64 |
| 32 | Render Output Units (ROPs) | 32 |
| 128 | Compute Units (Pipelines) | 128 |
| — | Tensor Cores | — |
| — | Ray-tracing Cores | 8 |
| — | L1 Cache | 1536KB shared |
| — | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |