Quick/Comparisons
14 | Gaming | 27 |
23 | Workstation | 32 |
10 | AI/ML |
Compare with something else?
Compare Apple M3 GPU with another →26 |
95 | Energy Efficiency | 61 |
30 | Quick Comparison Final Score | 34 |
| — | Graphics Processor | Navi 23 |
| 1280 | Cores | 2048 |
| 80 / 40 | TMUs / ROPs | 128 / 64 |
| 155.27 | Blender GPU | 1008.4 |
| — | Forza Horizon 5 | 65 |
| — | The Witcher 3 | 86 |
| — | Counter-Strike 2 | 118 |
| — | Far Cry 6 | 80 |
| — | Hogwarts Legacy | 53 |
| — | Call of Duty | 76 |
| — | Ghost of Tsushima | 54 |
| — | Cyberpunk 2077 | 56 |
| — | Tomb Raider | 96 |
| — | Average FPS | 76 |
| — | 1080p High | 98 |
| — | 1080p Ultra | 76 |
| — | 1440p Ultra | 53 |
| — | 4K Ultra | 28 |
| — | Margin of Error | Medium |
| 30027 | GB6 Compute Score | 79191 |
| 60.3 img/sec | Background Blur | 191.3 img/sec |
| 40.8 img/sec | Face Detection | 103.9 img/sec |
| 1.21 Gpixels/sec | Horizon Detection | 2.67 Gpixels/sec |
| 1.64 Gpixels/sec | Edge Detection | 3.6 Gpixels/sec |
| 1.2 Gpixels/sec | Gaussian Blur | 3.27 Gpixels/sec |
| 0.37 Gpixels/sec | Feature Matching | 0.88 Gpixels/sec |
| 102.6 Gpixels/sec | Stereo Matching | 278.7 Gpixels/sec |
| 3293.3 FPS | Particle Physics | 11307.3 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 15765 |
| — | G2D Mark | 729 |
| — | DirectX 11 | 138 FPS |
| — | DirectX 12 | 65 FPS |
| — | GPU Compute | 6043 Ops/s |
| 4120 | GB6 ML Single Precision | 14914 |
| 6992 | GB6 ML Half Precision | 16318 |
| 5638 | GB6 ML Quantized | 11595 |
| 1579 | Image Classification (SP) | 6441 |
| 2944 | Image Segmentation (HP) | 7535 |
| 4425 | Image Super Resolution (Q) | 17611 |
| 6661 | Face Detection (HP) | 21290 |
| 15020 | Pose Estimation (Q) | 69498 |
| 2984 | Text Classification (SP) | 2231 |
| 5876 | Machine Translation (HP) | 1967 |
| 1826 | Object Detection (SP) | 5939 |
| 9339 | Depth Estimation (Q) | 29299 |
| 22291 | Style Transfer (SP) | 155682 |
| Core ML | Framework | ONNX |
| CPU | Backend | DirectML |
| 55 GPixel/s | Pixel Fill Rate | 126 GPixel/s |
| 110 GTexel/s | Texture Fill Rate | 253 GTexel/s |
| 3.5 TFLOPS | FLOPS (FP32) | 8.1 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 x8 |
| 15 W | TGP | 100 W |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 7 nm |
| — | Die Size | 237 mm² |
| 13.8 billion | Transistor Count | 11.06 billion |
| — | Transistor Density | 46.67 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.3 |
| — | OpenGL | 4.6 |
| — | OpenCL | 2.1 |
| — | CUDA | — |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| — | DisplayPort | 1.4 |
| Apple | Vendor | Amd |
| Integrated | Build | Discrete |
| October 31, 2023 | Released | January 4, 2022 |
| Laptop | Case | Laptop |
| Professional | Purpose | Gaming |
| Mid-range | Segment | High-end |
| Apple M GPU | Architecture | RDNA 2.0 |
| — | GPU Codename | Navi 23 |
| -Apple M3or above | Recommended CPU | -AMD Ryzen 9 6900HSor above |
| 3478 | Steel Nomad Lite Score | 7812 |
| — | Time Spy | 8724 |
| 13529 | Solar Bay | 33935 |
| — | Port Royal | 4290 |
| — | Fire Strike | 26163 |
| 8229 | Wild Life Extreme | 16089 |
| — | Night Raid | 90630 |
| 500 MHz | Base Clock | 1800 MHz |
| 1380 MHz | Boost Clock | 1975 MHz |
| 1280 | Shading Units | 2048 |
| 80 | Texture Mapping Units (TMUs) | 128 |
| 40 | Render Output Units (ROPs) | 64 |
| 160 | Compute Units (Pipelines) | 32 |
| — | Tensor Cores | — |
| — | Ray-tracing Cores | 32 |
| — | L1 Cache | 128KB per cluster |
| — | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |