Quick/Comparisons
15 | Gaming | 23 |
25 | Workstation | 29 |
21 | AI/ML |
Compare with something else?
Compare Apple M4 GPU with another →13 |
95 | Energy Efficiency | 28 |
32 | Quick Comparison Final Score | 25 |
| Custom | Graphics Processor | GP104 |
| 1280 | Cores | 2560 |
| 80 / 40 | TMUs / ROPs | 160 / 64 |
| 1062.09 | Blender GPU | 585.53 |
| — | Tomb Raider | 81 |
| — | Average FPS | 66 |
| — | Forza Horizon 5 | 60 |
| — | The Witcher 3 | 79 |
| — | Counter-Strike 2 | 112 |
| — | Far Cry 6 | 71 |
| — | Hogwarts Legacy | 42 |
| — | Call of Duty | 56 |
| — | Ghost of Tsushima | 41 |
| — | Cyberpunk 2077 | 49 |
| — | 1080p High | 83 |
| — | 1080p Ultra | 66 |
| — | 1440p Ultra | 47 |
| — | 4K Ultra | 25 |
| — | Margin of Error | Low |
| 37678 | GB6 Compute Score | 59290 |
| 72.2 img/sec | Background Blur | 151.7 img/sec |
| 48.3 img/sec | Face Detection | 71.7 img/sec |
| 1.4 Gpixels/sec | Horizon Detection | 2.51 Gpixels/sec |
| 1.94 Gpixels/sec | Edge Detection | 3.71 Gpixels/sec |
| 1.5 Gpixels/sec | Gaussian Blur | 4.63 Gpixels/sec |
| 0.53 Gpixels/sec | Feature Matching | 0.29 Gpixels/sec |
| 123.3 Gpixels/sec | Stereo Matching | 214.4 Gpixels/sec |
| 4938.4 FPS | Particle Physics | 5989.1 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 11646 |
| — | G2D Mark | 722 |
| — | DirectX 11 | 101 FPS |
| — | DirectX 12 | 42 FPS |
| — | GPU Compute | 5844 Ops/s |
| 10346 | GB6 ML Single Precision | 6985 |
| 11883 | GB6 ML Half Precision | 6791 |
| 11036 | GB6 ML Quantized | 5167 |
| 4874 | Image Classification (SP) | 1863 |
| 8168 | Image Segmentation (HP) | 3814 |
| 11983 | Image Super Resolution (Q) | 9767 |
| 17036 | Face Detection (HP) | 6946 |
| 31171 | Pose Estimation (Q) | 49863 |
| 2964 | Text Classification (SP) | 1259 |
| 3410 | Machine Translation (HP) | 1365 |
| 5112 | Object Detection (SP) | 2518 |
| 22700 | Depth Estimation (Q) | 13518 |
| 60872 | Style Transfer (SP) | 99268 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 59 GPixel/s | Pixel Fill Rate | 111 GPixel/s |
| 118 GTexel/s | Texture Fill Rate | 277 GTexel/s |
| 3.8 TFLOPS | FLOPS (FP32) | 8.9 TFLOPS |
| System Shared | Memory Type | GDDR5X |
| — | Memory Size | 8 GB |
| 7500 MHz | Memory Clock | 1251 MHz |
| 15000 Mbps | Effective Memory Speed | 10008 Mbps |
| 128-bit | Bus | 256-bit |
| No | ECC | No |
| 120 GB/s | Memory Bandwidth | 320.3 GB/s |
| Custom | Interface | PCIe 3.0 x16 |
| 18 W | TGP | 150 W |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 16 nm |
| — | Die Size | 314 mm² |
| 28 billion | Transistor Count | 7.2 billion |
| — | Transistor Density | 22.93 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.3 |
| — | OpenGL | 4.6 |
| — | OpenCL | 1.2 |
| — | CUDA | 6.1 |
| Yes | Ray Tracing | No |
| No | DLSS | No |
| — | DisplayPort | 1.4 |
| Apple | Vendor | Nvidia |
| Integrated | Build | Discrete |
| May 7, 2024 | Released | August 15, 2016 |
| Laptop | Case | Laptop |
| Professional | Purpose | Gaming |
| Mid-range | Segment | High-end |
| Apple M GPU | Architecture | Pascal |
| Custom | GPU Codename | GP104 |
| -Apple M4 (10-Core)or above | Recommended CPU | -Intel Core i7 8750Hor above |
| 4097 | Steel Nomad Lite Score | 6942 |
| — | Time Spy | 7137 |
| 16545 | Solar Bay | 19261 |
| — | Port Royal | 1558 |
| — | Fire Strike | 21460 |
| 9591 | Wild Life Extreme | 13913 |
| — | Night Raid | 69100 |
| 500 MHz | Base Clock | 1556 MHz |
| 1470 MHz | Boost Clock | 1733 MHz |
| 1280 | Shading Units | 2560 |
| 80 | Texture Mapping Units (TMUs) | 160 |
| 40 | Render Output Units (ROPs) | 64 |
| 160 | Compute Units (Pipelines) | 20 |
| — | Tensor Cores | No |
| — | Ray-tracing Cores | No |
| — | L1 Cache | 48KB per cluster |
| — | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |