Quick/Comparisons
15 | Gaming | 7 |
25 | Workstation | 13 |
21 | AI/ML |
Compare with something else?
Compare Apple M4 GPU with another →7 |
95 | Energy Efficiency | 88 |
32 | Quick Comparison Final Score | 22 |
| Custom | Graphics Processor | Rembrandt |
| 1280 | Cores | 384 |
| 80 / 40 | TMUs / ROPs | 24 / 16 |
| 1062.09 | Blender GPU | 92.56 |
| — | Tomb Raider | 13 |
| — | Average FPS | 13 |
| — | Forza Horizon 5 | 12 |
| — | The Witcher 3 | 10 |
| — | Counter-Strike 2 | 20 |
| — | Far Cry 6 | 19 |
| — | Hogwarts Legacy | - |
| — | Call of Duty | 15 |
| — | Ghost of Tsushima | 9 |
| — | Cyberpunk 2077 | 9 |
| — | 1080p High | 17 |
| — | 1080p Ultra | 13 |
| — | 1440p Ultra | 9 |
| — | 4K Ultra | - |
| — | Margin of Error | Low |
| 37678 | GB6 Compute Score | 14065 |
| 72.2 img/sec | Background Blur | 36 img/sec |
| 48.3 img/sec | Face Detection | 16.4 img/sec |
| 1.4 Gpixels/sec | Horizon Detection | 0.48 Gpixels/sec |
| 1.94 Gpixels/sec | Edge Detection | 0.65 Gpixels/sec |
| 1.5 Gpixels/sec | Gaussian Blur | 0.53 Gpixels/sec |
| 0.53 Gpixels/sec | Feature Matching | 0.18 Gpixels/sec |
| 123.3 Gpixels/sec | Stereo Matching | 46.5 Gpixels/sec |
| 4938.4 FPS | Particle Physics | 2056.5 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 3207 |
| — | G2D Mark | 645 |
| — | DirectX 11 | 54 FPS |
| — | DirectX 12 | 27 FPS |
| — | GPU Compute | 1418 Ops/s |
| 10346 | GB6 ML Single Precision | 3903 |
| 11883 | GB6 ML Half Precision | 4745 |
| 11036 | GB6 ML Quantized | 2977 |
| 4874 | Image Classification (SP) | 1465 |
| 8168 | Image Segmentation (HP) | 1706 |
| 11983 | Image Super Resolution (Q) | 4266 |
| 17036 | Face Detection (HP) | 5115 |
| 31171 | Pose Estimation (Q) | 12992 |
| 2964 | Text Classification (SP) | 1116 |
| 3410 | Machine Translation (HP) | 1859 |
| 5112 | Object Detection (SP) | 1574 |
| 22700 | Depth Estimation (Q) | 6538 |
| 60872 | Style Transfer (SP) | 26656 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 59 GPixel/s | Pixel Fill Rate | 29-30 GPixel/s |
| 118 GTexel/s | Texture Fill Rate | 43-46 GTexel/s |
| 3.8 TFLOPS | FLOPS (FP32) | 1.4-1.5 TFLOPS |
| System Shared | Memory Type | System Shared |
| — | Memory Size | — |
| 7500 MHz | Memory Clock | — |
| 15000 Mbps | Effective Memory Speed | — |
| 128-bit | Bus | — |
| No | ECC | No |
| 120 GB/s | Memory Bandwidth | — |
| Custom | Interface | PCIe 4.0 x8 |
| 18 W | TGP | 15 W |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 6 nm |
| — | Die Size | 208 mm² |
| 28 billion | Transistor Count | 13.1 billion |
| — | Transistor Density | 62.98 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.2 |
| — | OpenGL | 4.6 |
| — | OpenCL | 2 |
| — | CUDA | — |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| — | DisplayPort | — |
| Apple | Vendor | Amd |
| Integrated | Build | Integrated |
| May 7, 2024 | Released | January 1, 2022 |
| Laptop | Case | Laptop |
| Professional | Purpose | Generic |
| Mid-range | Segment | Entry-level |
| Apple M GPU | Architecture | RDNA 2.0 |
| Custom | GPU Codename | Rembrandt |
| -Apple M4 (10-Core)or above | Recommended CPU | -AMD Ryzen 5 6600Hor above |
| 4097 | Steel Nomad Lite Score | 1344 |
| — | Time Spy | 1515 |
| 16545 | Solar Bay | 6396 |
| — | Port Royal | 586 |
| — | Fire Strike | 4870 |
| 9591 | Wild Life Extreme | 2676 |
| — | Night Raid | 19259 |
| 500 MHz | Base Clock | 400 MHz |
| 1470 MHz | Boost Clock | 1800-1900 MHz (configurable) |
| 1280 | Shading Units | 384 |
| 80 | Texture Mapping Units (TMUs) | 24 |
| 40 | Render Output Units (ROPs) | 16 |
| 160 | Compute Units (Pipelines) | 6 |
| — | Tensor Cores | — |
| — | Ray-tracing Cores | 6 |
| — | L1 Cache | 128KB per cluster |
| — | L2 Cache | 1024KB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |