Quick/Comparisons
14 | Gaming | 33 |
23 | Workstation | 42 |
10 | AI/ML |
Compare with something else?
Compare Apple M3 GPU with another →39 |
95 | Energy Efficiency | 82 |
30 | Quick Comparison Final Score | 44 |
| — | Graphics Processor | — |
| 1280 | Cores | 5120 |
| 80 / 40 | TMUs / ROPs | 320 / 160 |
| 155.27 | Blender GPU | 3797.82 |
| — | 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 | — |
| 30027 | GB6 Compute Score | 94528 |
| 60.3 img/sec | Background Blur | 157.6 img/sec |
| 40.8 img/sec | Face Detection | 100.9 img/sec |
| 1.21 Gpixels/sec | Horizon Detection | 3.99 Gpixels/sec |
| 1.64 Gpixels/sec | Edge Detection | 6.04 Gpixels/sec |
| 1.2 Gpixels/sec | Gaussian Blur | 5 Gpixels/sec |
| 0.37 Gpixels/sec | Feature Matching | 0.84 Gpixels/sec |
| 102.6 Gpixels/sec | Stereo Matching | 363.8 Gpixels/sec |
| 3293.3 FPS | Particle Physics | 12226.9 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | — |
| — | G2D Mark | — |
| — | DirectX 11 | — |
| — | DirectX 12 | — |
| — | GPU Compute | — |
| 4120 | GB6 ML Single Precision | 20068 |
| 6992 | GB6 ML Half Precision | 22390 |
| 5638 | GB6 ML Quantized | 20681 |
| 1579 | Image Classification (SP) | 8685 |
| 2944 | Image Segmentation (HP) | 20739 |
| 4425 | Image Super Resolution (Q) | 22775 |
| 6661 | Face Detection (HP) | 37676 |
| 15020 | Pose Estimation (Q) | 91609 |
| 2984 | Text Classification (SP) | 2909 |
| 5876 | Machine Translation (HP) | 5274 |
| 1826 | Object Detection (SP) | 7480 |
| 9339 | Depth Estimation (Q) | 38212 |
| 22291 | Style Transfer (SP) | 190681 |
| Core ML | Framework | Core ML |
| CPU | Backend | GPU |
| 55 GPixel/s | Pixel Fill Rate | 221 GPixel/s |
| 110 GTexel/s | Texture Fill Rate | 442 GTexel/s |
| 3.5 TFLOPS | FLOPS (FP32) | 14.1 TFLOPS |
| System Shared | Memory Type | System Shared |
| — | Memory Size | — |
| 6400 MHz | Memory Clock | 6400 MHz |
| 12800 Mbps | Effective Memory Speed | 12800 Mbps |
| 128-bit | Bus | 512-bit |
| No | ECC | No |
| 102.4 GB/s | Memory Bandwidth | 409.6 GB/s |
| Custom | Interface | Custom |
| 15 W | TGP | 60 W |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 3 nm |
| — | Die Size | — |
| 13.8 billion | Transistor Count | 56 billion |
| — | Transistor Density | — |
| — | DirectX | — |
| — | Vulkan | — |
| — | OpenGL | — |
| — | OpenCL | — |
| — | CUDA | — |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| — | DisplayPort | — |
| Apple | Vendor | Apple |
| Integrated | Build | Integrated |
| October 31, 2023 | Released | October 31, 2023 |
| Laptop | Case | Laptop |
| Professional | Purpose | Professional |
| Mid-range | Segment | High-end |
| Apple M GPU | Architecture | Apple M GPU |
| — | GPU Codename | — |
| -Apple M3or above | Recommended CPU | -Apple M3 Maxor above |
| 3478 | Steel Nomad Lite Score | 12391 |
| — | Time Spy | — |
| 13529 | Solar Bay | 49656 |
| — | Port Royal | — |
| — | Fire Strike | — |
| 8229 | Wild Life Extreme | 30787 |
| — | Night Raid | — |
| 500 MHz | Base Clock | 500 MHz |
| 1380 MHz | Boost Clock | 1380 MHz |
| 1280 | Shading Units | 5120 |
| 80 | Texture Mapping Units (TMUs) | 320 |
| 40 | Render Output Units (ROPs) | 160 |
| 160 | Compute Units (Pipelines) | 640 |
| — | Tensor Cores | — |
| — | Ray-tracing Cores | — |
| — | L1 Cache | — |
| — | L2 Cache | — |
| 2 IPC | Instructions Per Cycle | 2 IPC |