Quick/Comparisons
14 | Gaming | 16 |
23 | Workstation | 22 |
10 | AI/ML |
Compare with something else?
Compare Apple M3 GPU with another →17 |
95 | Energy Efficiency | 61 |
30 | Quick Comparison Final Score | 19 |
| — | Graphics Processor | Navi 24 |
| 1280 | Cores | 1024 |
| 80 / 40 | TMUs / ROPs | 64 / 32 |
| 155.27 | Blender GPU | 422.78 |
| — | Forza Horizon 5 | 43 |
| — | The Witcher 3 | 49 |
| — | Counter-Strike 2 | 69 |
| — | Far Cry 6 | 54 |
| — | Hogwarts Legacy | 27 |
| — | Call of Duty | 45 |
| — | Ghost of Tsushima | 29 |
| — | Cyberpunk 2077 | 34 |
| — | Tomb Raider | 59 |
| — | Average FPS | 45 |
| — | 1080p High | 59 |
| — | 1080p Ultra | 45 |
| — | 1440p Ultra | 29 |
| — | 4K Ultra | 15 |
| — | Margin of Error | Low |
| 30027 | GB6 Compute Score | 53081 |
| 60.3 img/sec | Background Blur | 130.3 img/sec |
| 40.8 img/sec | Face Detection | 70.1 img/sec |
| 1.21 Gpixels/sec | Horizon Detection | 1.59 Gpixels/sec |
| 1.64 Gpixels/sec | Edge Detection | 2.21 Gpixels/sec |
| 1.2 Gpixels/sec | Gaussian Blur | 2.02 Gpixels/sec |
| 0.37 Gpixels/sec | Feature Matching | 0.65 Gpixels/sec |
| 102.6 Gpixels/sec | Stereo Matching | 223 Gpixels/sec |
| 3293.3 FPS | Particle Physics | 7470.5 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 9632 |
| — | G2D Mark | 817 |
| — | DirectX 11 | 90 FPS |
| — | DirectX 12 | 38 FPS |
| — | GPU Compute | 4342 Ops/s |
| 4120 | GB6 ML Single Precision | 10359 |
| 6992 | GB6 ML Half Precision | 12861 |
| 5638 | GB6 ML Quantized | 7842 |
| 1579 | Image Classification (SP) | 3867 |
| 2944 | Image Segmentation (HP) | 5136 |
| 4425 | Image Super Resolution (Q) | 11758 |
| 6661 | Face Detection (HP) | 14844 |
| 15020 | Pose Estimation (Q) | 42787 |
| 2984 | Text Classification (SP) | 1986 |
| 5876 | Machine Translation (HP) | 2665 |
| 1826 | Object Detection (SP) | 4251 |
| 9339 | Depth Estimation (Q) | 19707 |
| 22291 | Style Transfer (SP) | 101802 |
| Core ML | Framework | ONNX |
| CPU | Backend | DirectML |
| 55 GPixel/s | Pixel Fill Rate | 90 GPixel/s |
| 110 GTexel/s | Texture Fill Rate | 180 GTexel/s |
| 3.5 TFLOPS | FLOPS (FP32) | 5.8 TFLOPS |
| System Shared | Memory Type | GDDR6 |
| — | Memory Size | 4 GB |
| 6400 MHz | Memory Clock | 2248 MHz |
| 12800 Mbps | Effective Memory Speed | 18000 Mbps |
| 128-bit | Bus | 64-bit |
| No | ECC | No |
| 102.4 GB/s | Memory Bandwidth | 143.9 GB/s |
| Custom | Interface | PCIe 4.0 x4 |
| 15 W | TGP | 107 W |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 6 nm |
| — | Die Size | 107 mm² |
| 13.8 billion | Transistor Count | 5 billion |
| — | Transistor Density | 46.73 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.3 |
| — | OpenGL | 4.6 |
| — | OpenCL | 2.2 |
| — | CUDA | — |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| — | DisplayPort | 1.4a |
| Apple | Vendor | Amd |
| Integrated | Build | Discrete |
| October 31, 2023 | Released | January 19, 2022 |
| Laptop | Case | Desktop |
| Professional | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Apple M GPU | Architecture | RDNA 2.0 |
| — | GPU Codename | Navi 24 |
| -Apple M3or above | Recommended CPU | -Intel Core i5 12400or above |
| 3478 | Steel Nomad Lite Score | 4764 |
| — | Time Spy | 4943 |
| 13529 | Solar Bay | 20002 |
| — | Port Royal | 368 |
| — | Fire Strike | 15575 |
| 8229 | Wild Life Extreme | 8989 |
| — | Night Raid | 65384 |
| 500 MHz | Base Clock | 2310 MHz |
| 1380 MHz | Boost Clock | 2815 MHz |
| 1280 | Shading Units | 1024 |
| 80 | Texture Mapping Units (TMUs) | 64 |
| 40 | Render Output Units (ROPs) | 32 |
| 160 | Compute Units (Pipelines) | — |
| — | Tensor Cores | No |
| — | Ray-tracing Cores | 16 |
| — | L1 Cache | 128KB per cluster |
| — | L2 Cache | 1024KB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |