Quick/Comparisons
9 | Gaming | 34 |
23 | Workstation | 36 |
12 | AI/ML |
Compare with something else?
Compare Apple M2 GPU with another →31 |
91 | Energy Efficiency | 56 |
27 | Quick Comparison Final Score | 38 |
| — | Graphics Processor | AD107 (GN21-X4) |
| 1024 | Cores | 3072 |
| 64 / 32 | TMUs / ROPs | 96 / 48 |
| 301.33 | Blender GPU | 3309.28 |
| — | Forza Horizon 5 | 76 |
| — | The Witcher 3 | 110 |
| — | Counter-Strike 2 | 150 |
| — | Far Cry 6 | 96 |
| — | Hogwarts Legacy | 63 |
| — | Call of Duty | 86 |
| — | Ghost of Tsushima | 58 |
| — | Cyberpunk 2077 | 73 |
| — | Tomb Raider | 116 |
| — | Average FPS | 92 |
| — | 1080p High | 114 |
| — | 1080p Ultra | 92 |
| — | 1440p Ultra | 67 |
| — | 4K Ultra | 37 |
| — | Margin of Error | Low |
| 29705 | GB6 Compute Score | 97986 |
| 58.7 img/sec | Background Blur | 142.7 img/sec |
| 35.2 img/sec | Face Detection | 103.7 img/sec |
| 1.1 Gpixels/sec | Horizon Detection | 3.83 Gpixels/sec |
| 1.82 Gpixels/sec | Edge Detection | 4.93 Gpixels/sec |
| 1.2 Gpixels/sec | Gaussian Blur | 4.63 Gpixels/sec |
| 0.37 Gpixels/sec | Feature Matching | 1.27 Gpixels/sec |
| 96 Gpixels/sec | Stereo Matching | 430.4 Gpixels/sec |
| 3825.9 FPS | Particle Physics | 13519.5 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 17398 |
| — | G2D Mark | 731 |
| — | DirectX 11 | 156 FPS |
| — | DirectX 12 | 72 FPS |
| — | GPU Compute | 6811 Ops/s |
| 6206 | GB6 ML Single Precision | 17683 |
| 6893 | GB6 ML Half Precision | 30738 |
| 6515 | GB6 ML Quantized | 13612 |
| 3242 | Image Classification (SP) | 7421 |
| 4890 | Image Segmentation (HP) | 19041 |
| 7149 | Image Super Resolution (Q) | 20913 |
| 10340 | Face Detection (HP) | 35464 |
| 21268 | Pose Estimation (Q) | 67110 |
| 2268 | Text Classification (SP) | 3086 |
| 1122 | Machine Translation (HP) | 4740 |
| 3007 | Object Detection (SP) | 8644 |
| 13941 | Depth Estimation (Q) | 32456 |
| 35943 | Style Transfer (SP) | 142039 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 45 GPixel/s | Pixel Fill Rate | 71-114 GPixel/s |
| 89 GTexel/s | Texture Fill Rate | 141-228 GTexel/s |
| 2.9 TFLOPS | FLOPS (FP32) | 9-14.6 TFLOPS |
| System Shared | Memory Type | GDDR6 |
| — | Memory Size | 8 GB |
| 6400 MHz | Memory Clock | 2000 MHz |
| 12800 Mbps | Effective Memory Speed | 16000 Mbps |
| 128-bit | Bus | 128-bit |
| No | ECC | No |
| 102.4 GB/s | Memory Bandwidth | 256 GB/s |
| Custom | Interface | PCIe 4.0 x16 |
| 15 W | TGP | 35-140 W (configurable) |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 5 nm |
| — | Die Size | 156 mm² |
| 11.6 billion | Transistor Count | 18.9 billion |
| — | Transistor Density | 121.15 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.3 |
| — | OpenGL | 4.6 |
| — | OpenCL | 3 |
| — | CUDA | 8.9 |
| No | Ray Tracing | Yes |
| No | DLSS | DLSS 3 |
| — | DisplayPort | 1.4a |
| Apple | Vendor | Nvidia |
| Integrated | Build | Discrete |
| June 6, 2022 | Released | February 22, 2023 |
| Laptop | Case | Laptop |
| Professional | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Apple M GPU | Architecture | Ada Lovelace |
| — | GPU Codename | AD107 (GN21-X4) |
| -Apple M2or above | Recommended CPU | -Intel Core Ultra 7 155Hor above |
| 2312 | Steel Nomad Lite Score | 10224 |
| — | Time Spy | 10403 |
| 7310 | Solar Bay | 44947 |
| — | Port Royal | 5819 |
| — | Fire Strike | 26488 |
| 5675 | Wild Life Extreme | 19808 |
| — | Night Raid | 98392 |
| 500 MHz | Base Clock | 1140-2295 MHz (configurable) |
| 1398 MHz | Boost Clock | 1470-2370 MHz (configurable) |
| 1024 | Shading Units | 3072 |
| 64 | Texture Mapping Units (TMUs) | 96 |
| 32 | Render Output Units (ROPs) | 48 |
| 128 | Compute Units (Pipelines) | 24 |
| — | Tensor Cores | 96 |
| — | Ray-tracing Cores | 24 |
| — | L1 Cache | 128KB per cluster |
| — | L2 Cache | 32MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |