Quick/Comparisons
8 | Gaming | 34 |
21 | Workstation | 41 |
7 | AI/ML |
Compare with something else?
Compare Apple M1 GPU with another →31 |
89 | Energy Efficiency | 43 |
25 | Quick Comparison Final Score | 37 |
| — | Graphics Processor | GA104 (GN20-E5) |
| 1024 | Cores | 5120 |
| 64 / 32 | TMUs / ROPs | 160 / 80 |
| 259.48 | Blender GPU | 2998.92 |
| — | Forza Horizon 5 | 79 |
| — | The Witcher 3 | 114 |
| — | Counter-Strike 2 | 150 |
| — | Far Cry 6 | 94 |
| — | Hogwarts Legacy | 61 |
| — | Call of Duty | 88 |
| — | Ghost of Tsushima | 56 |
| — | Cyberpunk 2077 | 73 |
| — | Tomb Raider | 116 |
| — | Average FPS | 92 |
| — | 1080p High | 114 |
| — | 1080p Ultra | 92 |
| — | 1440p Ultra | 68 |
| — | 4K Ultra | 38 |
| — | Margin of Error | Low |
| 20842 | GB6 Compute Score | 118499 |
| 38.3 img/sec | Background Blur | 170.3 img/sec |
| 23 img/sec | Face Detection | 113.6 img/sec |
| 0.76 Gpixels/sec | Horizon Detection | 5.59 Gpixels/sec |
| 1.33 Gpixels/sec | Edge Detection | 7.94 Gpixels/sec |
| 0.87 Gpixels/sec | Gaussian Blur | 6.19 Gpixels/sec |
| 0.27 Gpixels/sec | Feature Matching | 1.16 Gpixels/sec |
| 69.9 Gpixels/sec | Stereo Matching | 467 Gpixels/sec |
| 2766.4 FPS | Particle Physics | 15131.3 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 15105 |
| — | G2D Mark | 630 |
| — | DirectX 11 | 137 FPS |
| — | DirectX 12 | 63 FPS |
| — | GPU Compute | 6702 Ops/s |
| 3284 | GB6 ML Single Precision | 17250 |
| 5418 | GB6 ML Half Precision | 27936 |
| 4395 | GB6 ML Quantized | 11249 |
| 1400 | Image Classification (SP) | 6969 |
| 2280 | Image Segmentation (HP) | 17142 |
| 3402 | Image Super Resolution (Q) | 16426 |
| 5638 | Face Detection (HP) | 31628 |
| 12027 | Pose Estimation (Q) | 74413 |
| 2324 | Text Classification (SP) | 2522 |
| 4296 | Machine Translation (HP) | 3639 |
| 1530 | Object Detection (SP) | 9002 |
| 7675 | Depth Estimation (Q) | 27790 |
| 17293 | Style Transfer (SP) | 160005 |
| Core ML | Framework | ONNX |
| CPU | Backend | DirectML |
| 41 GPixel/s | Pixel Fill Rate | 103-130 GPixel/s |
| 82 GTexel/s | Texture Fill Rate | 206-259 GTexel/s |
| 2.6 TFLOPS | FLOPS (FP32) | 13.2-16.6 TFLOPS |
| System Shared | Memory Type | GDDR6 |
| — | Memory Size | 8 GB |
| 4266 MHz | Memory Clock | 1500-1750 MHz |
| 8532 Mbps | Effective Memory Speed | 12000-14000 Mbps |
| 128-bit | Bus | 256-bit |
| No | ECC | No |
| 68.2 GB/s | Memory Bandwidth | 448 GB/s |
| Custom | Interface | PCIe 4.0 x16 |
| 15 W | TGP | 80-150 W (configurable) |
| Intel | Manufacturing | Samsung |
| 5 nm | Fabrication Process | 8 nm |
| — | Die Size | 392 mm² |
| 11.6 billion | Transistor Count | 17.4 billion |
| — | Transistor Density | 44.39 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.2 |
| — | OpenGL | 4.6 |
| — | OpenCL | 2 |
| — | CUDA | 8.6 |
| No | Ray Tracing | Yes |
| No | DLSS | DLSS 2 |
| — | DisplayPort | 1.4a |
| Apple | Vendor | Nvidia |
| Integrated | Build | Discrete |
| November 20, 2020 | Released | January 12, 2021 |
| Laptop | Case | Laptop |
| Professional | Purpose | Gaming |
| Mid-range | Segment | High-end |
| Apple M GPU | Architecture | Ampere |
| — | GPU Codename | GA104 (GN20-E5) |
| -Apple M1or above | Recommended CPU | -Intel Core i7 12700Hor above |
| 1979 | Steel Nomad Lite Score | 10214 |
| — | Time Spy | 10477 |
| 6311 | Solar Bay | 45215 |
| — | Port Royal | 6247 |
| — | Fire Strike | 26729 |
| 4957 | Wild Life Extreme | 21137 |
| — | Night Raid | 82533 |
| 450 MHz | Base Clock | 780-1215 MHz (configurable) |
| 1278 MHz | Boost Clock | 1290-1620 MHz (configurable) |
| 1024 | Shading Units | 5120 |
| 64 | Texture Mapping Units (TMUs) | 160 |
| 32 | Render Output Units (ROPs) | 80 |
| 128 | Compute Units (Pipelines) | 40 |
| — | Tensor Cores | 160 |
| No | Ray-tracing Cores | 40 |
| — | L1 Cache | 128KB per cluster |
| — | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |