Quick/Comparisons
19 | Gaming | 29 |
32 | Workstation | 36 |
34 | AI/ML |
Compare with something else?
Compare Apple M5 GPU with another →29 |
100 | Energy Efficiency | 28 |
38 | Quick Comparison Final Score | 31 |
| Custom | Graphics Processor | TU106 |
| 1280 | Cores | 2176 |
| 80 / 40 | TMUs / ROPs | 136 / 64 |
| 264.45 | Blender GPU | 2199.09 |
| — | Forza Horizon 5 | 65 |
| — | The Witcher 3 | 96 |
| — | Counter-Strike 2 | 128 |
| — | Far Cry 6 | 81 |
| — | Hogwarts Legacy | 55 |
| — | Call of Duty | 68 |
| — | Ghost of Tsushima | 50 |
| — | Cyberpunk 2077 | 60 |
| — | Tomb Raider | 97 |
| — | Average FPS | 78 |
| — | 1080p High | 97 |
| — | 1080p Ultra | 78 |
| — | 1440p Ultra | 57 |
| — | 4K Ultra | 31 |
| — | Margin of Error | Low |
| 75777 | GB6 Compute Score | 92186 |
| — | Background Blur | 155.7 img/sec |
| — | Face Detection | 93.7 img/sec |
| — | Horizon Detection | 3.81 Gpixels/sec |
| — | Edge Detection | 4.91 Gpixels/sec |
| — | Gaussian Blur | 4.89 Gpixels/sec |
| — | Feature Matching | 1.09 Gpixels/sec |
| — | Stereo Matching | 381.7 Gpixels/sec |
| — | Particle Physics | 10550.4 FPS |
| — | API | OpenCL |
| — | G3D Mark Score | 16471 |
| — | G2D Mark | 857 |
| — | DirectX 11 | 131 FPS |
| — | DirectX 12 | 61 FPS |
| — | GPU Compute | 6686 Ops/s |
| 13289 | GB6 ML Single Precision | 13870 |
| 24800 | GB6 ML Half Precision | 24064 |
| 23572 | GB6 ML Quantized | 9639 |
| 6348 | Image Classification (SP) | 5952 |
| 16813 | Image Segmentation (HP) | 13581 |
| 30322 | Image Super Resolution (Q) | 12282 |
| 34387 | Face Detection (HP) | 27779 |
| 111591 | Pose Estimation (Q) | 53249 |
| 2319 | Text Classification (SP) | 2087 |
| 5555 | Machine Translation (HP) | 3660 |
| 6450 | Object Detection (SP) | 6977 |
| 53195 | Depth Estimation (Q) | 23707 |
| 82314 | Style Transfer (SP) | 112439 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 65 GPixel/s | Pixel Fill Rate | 106 GPixel/s |
| 130 GTexel/s | Texture Fill Rate | 224 GTexel/s |
| 4.1 TFLOPS | FLOPS (FP32) | 7.2 TFLOPS |
| System Shared | Memory Type | GDDR6 |
| — | Memory Size | 8 GB |
| 9600 MHz | Memory Clock | 1750 MHz |
| — | Effective Memory Speed | 14000 Mbps |
| 128-bit | Bus | 256-bit |
| No | ECC | No |
| 153 GB/s | Memory Bandwidth | 448 GB/s |
| Custom | Interface | PCIe 3.0 x16 |
| — | TGP | 175 W |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 12 nm |
| — | Die Size | 445 mm² |
| 28 billion | Transistor Count | 10 billion |
| — | Transistor Density | 22.47 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.3 |
| — | OpenGL | 4.6 |
| — | OpenCL | 3 |
| — | CUDA | 7.5 |
| Yes | Ray Tracing | Yes |
| No | DLSS | DLSS 2 |
| — | DisplayPort | 1.4a |
| Apple | Vendor | Nvidia |
| Integrated | Build | Discrete |
| October 15, 2025 | Released | July 9, 2019 |
| Laptop | Case | Desktop |
| Professional | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Apple M GPU | Architecture | Turing |
| Custom | GPU Codename | TU106 |
| -Apple M5 (10-Core)or above | Recommended CPU | -Intel Core i5 9600Kor above |
| 5337 | Steel Nomad Lite Score | 8311 |
| — | Time Spy | 8732 |
| 24534 | Solar Bay | 38547 |
| — | Port Royal | 5061 |
| — | Fire Strike | 22538 |
| 12266 | Wild Life Extreme | 17523 |
| — | Night Raid | 92195 |
| — | Base Clock | 1470 MHz |
| 1620 MHz | Boost Clock | 1650 MHz |
| 1280 | Shading Units | 2176 |
| 80 | Texture Mapping Units (TMUs) | 136 |
| 40 | Render Output Units (ROPs) | 64 |
| 160 | Compute Units (Pipelines) | 34 |
| — | Tensor Cores | 272 |
| — | Ray-tracing Cores | 34 |
| — | L1 Cache | 64KB per cluster |
| — | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |