Quick/Comparisons
15 | Gaming | 12 |
25 | Workstation | 17 |
21 | AI/ML |
Compare with something else?
Compare Apple M4 GPU with another →18 |
95 | Energy Efficiency | 85 |
32 | Quick Comparison Final Score | 26 |
| Custom | Graphics Processor | Strix Point |
| 1280 | Cores | 1024 |
| 80 / 40 | TMUs / ROPs | 64 / 32 |
| 1062.09 | Blender GPU | 311.81 |
| — | Tomb Raider | 35 |
| — | Average FPS | 28 |
| — | Forza Horizon 5 | 28 |
| — | The Witcher 3 | 28 |
| — | Counter-Strike 2 | 32 |
| — | Far Cry 6 | 38 |
| — | Hogwarts Legacy | 19 |
| — | Call of Duty | 32 |
| — | Ghost of Tsushima | 16 |
| — | Cyberpunk 2077 | 22 |
| — | 1080p High | 38 |
| — | 1080p Ultra | 28 |
| — | 1440p Ultra | 19 |
| — | 4K Ultra | 10 |
| — | Margin of Error | Low |
| 37678 | GB6 Compute Score | 42932 |
| 72.2 img/sec | Background Blur | 98.6 img/sec |
| 48.3 img/sec | Face Detection | 42 img/sec |
| 1.4 Gpixels/sec | Horizon Detection | 1.25 Gpixels/sec |
| 1.94 Gpixels/sec | Edge Detection | 1.6 Gpixels/sec |
| 1.5 Gpixels/sec | Gaussian Blur | 1.69 Gpixels/sec |
| 0.53 Gpixels/sec | Feature Matching | 0.55 Gpixels/sec |
| 123.3 Gpixels/sec | Stereo Matching | 214.8 Gpixels/sec |
| 4938.4 FPS | Particle Physics | 7757.7 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 8122 |
| — | G2D Mark | 976 |
| — | DirectX 11 | 73 FPS |
| — | DirectX 12 | 37 FPS |
| — | GPU Compute | 4164 Ops/s |
| 10346 | GB6 ML Single Precision | 9613 |
| 11883 | GB6 ML Half Precision | 15157 |
| 11036 | GB6 ML Quantized | 7082 |
| 4874 | Image Classification (SP) | 3563 |
| 8168 | Image Segmentation (HP) | 6170 |
| 11983 | Image Super Resolution (Q) | 10762 |
| 17036 | Face Detection (HP) | 13773 |
| 31171 | Pose Estimation (Q) | 37861 |
| 2964 | Text Classification (SP) | 2572 |
| 3410 | Machine Translation (HP) | 4699 |
| 5112 | Object Detection (SP) | 2845 |
| 22700 | Depth Estimation (Q) | 17260 |
| 60872 | Style Transfer (SP) | 86751 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 59 GPixel/s | Pixel Fill Rate | 93 GPixel/s |
| 118 GTexel/s | Texture Fill Rate | 186 GTexel/s |
| 3.8 TFLOPS | FLOPS (FP32) | 5.9 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 x16 |
| 18 W | TGP | 15-54 W (configurable) |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 4 nm |
| — | Die Size | 233 mm² |
| 28 billion | Transistor Count | 34 billion |
| — | Transistor Density | 145.92 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.4 |
| — | OpenGL | 4.6 |
| — | OpenCL | 2.1 |
| — | CUDA | None |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| — | DisplayPort | 2.1 |
| Apple | Vendor | Amd |
| Integrated | Build | Integrated |
| May 7, 2024 | Released | June 3, 2024 |
| Laptop | Case | Laptop |
| Professional | Purpose | Generic |
| Mid-range | Segment | Mid-range |
| Apple M GPU | Architecture | RDNA 3.5 |
| Custom | GPU Codename | Strix Point |
| -Apple M4 (10-Core)or above | Recommended CPU | -AMD Ryzen AI 9 HX 370or above |
| 4097 | Steel Nomad Lite Score | 3190 |
| — | Time Spy | 3307 |
| 16545 | Solar Bay | 15184 |
| — | Port Royal | 1790 |
| — | Fire Strike | 8885 |
| 9591 | Wild Life Extreme | 6775 |
| — | Night Raid | 37532 |
| 500 MHz | Base Clock | 400 MHz |
| 1470 MHz | Boost Clock | 2900 MHz |
| 1280 | Shading Units | 1024 |
| 80 | Texture Mapping Units (TMUs) | 64 |
| 40 | Render Output Units (ROPs) | 32 |
| 160 | Compute Units (Pipelines) | 16 |
| — | Tensor Cores | No |
| — | Ray-tracing Cores | 16 |
| — | L1 Cache | 512KB per cluster |
| — | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |