Quick/Comparisons
15 | Gaming | 12 |
25 | Workstation | 16 |
21 | AI/ML |
Compare with something else?
Compare Apple M4 GPU with another →18 |
95 | Energy Efficiency | 89 |
32 | Quick Comparison Final Score | 26 |
| Custom | Graphics Processor | Arrow Lake iGPU |
| 1280 | Cores | 1024 |
| 80 / 40 | TMUs / ROPs | 64 / 32 |
| 1062.09 | Blender GPU | 765.53 |
| — | Tomb Raider | 35 |
| — | Average FPS | 27 |
| — | Forza Horizon 5 | 29 |
| — | The Witcher 3 | 33 |
| — | Counter-Strike 2 | 33 |
| — | Far Cry 6 | 35 |
| — | Hogwarts Legacy | 14 |
| — | Call of Duty | 27 |
| — | Ghost of Tsushima | 17 |
| — | Cyberpunk 2077 | 20 |
| — | 1080p High | 35 |
| — | 1080p Ultra | 27 |
| — | 1440p Ultra | 19 |
| — | 4K Ultra | 10 |
| — | Margin of Error | Low |
| 37678 | GB6 Compute Score | 42502 |
| 72.2 img/sec | Background Blur | 90 img/sec |
| 48.3 img/sec | Face Detection | 39.5 img/sec |
| 1.4 Gpixels/sec | Horizon Detection | 1.21 Gpixels/sec |
| 1.94 Gpixels/sec | Edge Detection | 1.52 Gpixels/sec |
| 1.5 Gpixels/sec | Gaussian Blur | 2.41 Gpixels/sec |
| 0.53 Gpixels/sec | Feature Matching | 0.51 Gpixels/sec |
| 123.3 Gpixels/sec | Stereo Matching | 161.8 Gpixels/sec |
| 4938.4 FPS | Particle Physics | 9155.9 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 6620 |
| — | G2D Mark | 835 |
| — | DirectX 11 | 47 FPS |
| — | DirectX 12 | 29 FPS |
| — | GPU Compute | 2992 Ops/s |
| 10346 | GB6 ML Single Precision | 8764 |
| 11883 | GB6 ML Half Precision | 12046 |
| 11036 | GB6 ML Quantized | 5815 |
| 4874 | Image Classification (SP) | 5224 |
| 8168 | Image Segmentation (HP) | 10501 |
| 11983 | Image Super Resolution (Q) | 6436 |
| 17036 | Face Detection (HP) | 15277 |
| 31171 | Pose Estimation (Q) | 19008 |
| 2964 | Text Classification (SP) | 1566 |
| 3410 | Machine Translation (HP) | 2969 |
| 5112 | Object Detection (SP) | 6172 |
| 22700 | Depth Estimation (Q) | 14526 |
| 60872 | Style Transfer (SP) | 62049 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 59 GPixel/s | Pixel Fill Rate | 72-75 GPixel/s |
| 118 GTexel/s | Texture Fill Rate | 144-150 GTexel/s |
| 3.8 TFLOPS | FLOPS (FP32) | 4.6-4.8 TFLOPS |
| System Shared | Memory Type | System Shared |
| — | Memory Size | — |
| 7500 MHz | Memory Clock | — |
| 15000 Mbps | Effective Memory Speed | — |
| 128-bit | Bus | — |
| No | ECC | — |
| 120 GB/s | Memory Bandwidth | — |
| Custom | Interface | PCIe 4.0 x8 |
| 18 W | TGP | 35 W |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 5 nm |
| — | Die Size | — |
| 28 billion | Transistor Count | — |
| — | Transistor Density | — |
| — | DirectX | 12.2 |
| — | Vulkan | 1.3 |
| — | OpenGL | 4.6 |
| — | OpenCL | 3 |
| — | CUDA | No |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| — | DisplayPort | 2.1 |
| Apple | Vendor | Intel |
| Integrated | Build | Integrated |
| May 7, 2024 | Released | January 6, 2025 |
| Laptop | Case | Laptop |
| Professional | Purpose | Generic |
| Mid-range | Segment | Entry-level |
| Apple M GPU | Architecture | Xe+ |
| Custom | GPU Codename | Arrow Lake iGPU |
| -Apple M4 (10-Core)or above | Recommended CPU | -Intel Core Ultra 5 235Hor above |
| 4097 | Steel Nomad Lite Score | 3255 |
| — | Time Spy | 3793 |
| 16545 | Solar Bay | 14650 |
| — | Port Royal | 1990 |
| — | Fire Strike | 8598 |
| 9591 | Wild Life Extreme | 6648 |
| — | Night Raid | 38856 |
| 500 MHz | Base Clock | 300 MHz |
| 1470 MHz | Boost Clock | 2250-2350 MHz (configurable) |
| 1280 | Shading Units | 1024 |
| 80 | Texture Mapping Units (TMUs) | 64 |
| 40 | Render Output Units (ROPs) | 32 |
| 160 | Compute Units (Pipelines) | 8 |
| — | Tensor Cores | 128 |
| — | Ray-tracing Cores | 8 |
| — | L1 Cache | — |
| — | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |