Quick/Comparisons
15 | Gaming | 38 |
25 | Workstation | 42 |
21 | AI/ML |
Compare with something else?
Compare Apple M4 GPU with another →31 |
95 | Energy Efficiency | 35 |
32 | Quick Comparison Final Score | 38 |
| Custom | Graphics Processor | GA104 |
| 1280 | Cores | 4864 |
| 80 / 40 | TMUs / ROPs | 152 / 80 |
| 1062.09 | Blender GPU | 2992.04 |
| — | Tomb Raider | 128 |
| — | Average FPS | 104 |
| — | Forza Horizon 5 | 87 |
| — | The Witcher 3 | 124 |
| — | Counter-Strike 2 | 167 |
| — | Far Cry 6 | 106 |
| — | Hogwarts Legacy | 71 |
| — | Call of Duty | 100 |
| — | Ghost of Tsushima | 68 |
| — | Cyberpunk 2077 | 82 |
| — | 1080p High | 127 |
| — | 1080p Ultra | 104 |
| — | 1440p Ultra | 77 |
| — | 4K Ultra | 43 |
| — | Margin of Error | Low |
| 37678 | GB6 Compute Score | 115518 |
| 72.2 img/sec | Background Blur | 185.1 img/sec |
| 48.3 img/sec | Face Detection | 108.4 img/sec |
| 1.4 Gpixels/sec | Horizon Detection | 5.35 Gpixels/sec |
| 1.94 Gpixels/sec | Edge Detection | 7.17 Gpixels/sec |
| 1.5 Gpixels/sec | Gaussian Blur | 5.54 Gpixels/sec |
| 0.53 Gpixels/sec | Feature Matching | 1.15 Gpixels/sec |
| 123.3 Gpixels/sec | Stereo Matching | 478 Gpixels/sec |
| 4938.4 FPS | Particle Physics | 15183 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 20272 |
| — | G2D Mark | 990 |
| — | DirectX 11 | 164 FPS |
| — | DirectX 12 | 78 FPS |
| — | GPU Compute | 9905 Ops/s |
| 10346 | GB6 ML Single Precision | 17811 |
| 11883 | GB6 ML Half Precision | 29969 |
| 11036 | GB6 ML Quantized | 11367 |
| 4874 | Image Classification (SP) | 7384 |
| 8168 | Image Segmentation (HP) | 19320 |
| 11983 | Image Super Resolution (Q) | 16108 |
| 17036 | Face Detection (HP) | 34615 |
| 31171 | Pose Estimation (Q) | 74811 |
| 2964 | Text Classification (SP) | 2638 |
| 3410 | Machine Translation (HP) | 4136 |
| 5112 | Object Detection (SP) | 8718 |
| 22700 | Depth Estimation (Q) | 25946 |
| 60872 | Style Transfer (SP) | 153275 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 59 GPixel/s | Pixel Fill Rate | 133 GPixel/s |
| 118 GTexel/s | Texture Fill Rate | 253 GTexel/s |
| 3.8 TFLOPS | FLOPS (FP32) | 16.2 TFLOPS |
| System Shared | Memory Type | GDDR6 |
| — | Memory Size | 8 GB |
| 7500 MHz | Memory Clock | 1750 MHz |
| 15000 Mbps | Effective Memory Speed | 14000 Mbps |
| 128-bit | Bus | 256-bit |
| No | ECC | No |
| 120 GB/s | Memory Bandwidth | 448 GB/s |
| Custom | Interface | PCIe 4.0 x16 |
| 18 W | TGP | 200 W |
| TSMC | Manufacturing | Samsung |
| 3 nm | Fabrication Process | 8 nm |
| — | Die Size | 392 mm² |
| 28 billion | Transistor Count | 17 billion |
| — | Transistor Density | 43.37 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.3 |
| — | OpenGL | 4.6 |
| — | OpenCL | 3 |
| — | CUDA | 8.6 |
| Yes | Ray Tracing | Yes |
| No | DLSS | DLSS 2 |
| — | DisplayPort | 1.4a |
| Apple | Vendor | Nvidia |
| Integrated | Build | Discrete |
| May 7, 2024 | Released | December 2, 2020 |
| Laptop | Case | Desktop |
| Professional | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Apple M GPU | Architecture | Ampere |
| Custom | GPU Codename | GA104 |
| -Apple M4 (10-Core)or above | Recommended CPU | -Intel Core i5 12600Kor above |
| 4097 | Steel Nomad Lite Score | 11978 |
| — | Time Spy | 11707 |
| 16545 | Solar Bay | 54374 |
| — | Port Royal | 6971 |
| — | Fire Strike | 29568 |
| 9591 | Wild Life Extreme | 24415 |
| — | Night Raid | 109854 |
| 500 MHz | Base Clock | 1410 MHz |
| 1470 MHz | Boost Clock | 1665 MHz |
| 1280 | Shading Units | 4864 |
| 80 | Texture Mapping Units (TMUs) | 152 |
| 40 | Render Output Units (ROPs) | 80 |
| 160 | Compute Units (Pipelines) | 38 |
| — | Tensor Cores | 152 |
| — | Ray-tracing Cores | 38 |
| — | L1 Cache | 128KB per cluster |
| — | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |