Quick/Comparisons
9 | Gaming | 37 |
23 | Workstation | 42 |
12 | AI/ML |
Compare with something else?
Compare Apple M2 GPU with another →32 |
91 | Energy Efficiency | 27 |
27 | Quick Comparison Final Score | 38 |
| — | Graphics Processor | TU104 |
| 1024 | Cores | 3072 |
| 64 / 32 | TMUs / ROPs | 192 / 64 |
| 301.33 | Blender GPU | 2407.76 |
| — | Forza Horizon 5 | 86 |
| — | The Witcher 3 | 121 |
| — | Counter-Strike 2 | 169 |
| — | Far Cry 6 | 105 |
| — | Hogwarts Legacy | 68 |
| — | Call of Duty | 97 |
| — | Ghost of Tsushima | 64 |
| — | Cyberpunk 2077 | 82 |
| — | Tomb Raider | 131 |
| — | Average FPS | 103 |
| — | 1080p High | 124 |
| — | 1080p Ultra | 103 |
| — | 1440p Ultra | 75 |
| — | 4K Ultra | 42 |
| — | Margin of Error | Low |
| 29705 | GB6 Compute Score | 122905 |
| 58.7 img/sec | Background Blur | 197.1 img/sec |
| 35.2 img/sec | Face Detection | 110.7 img/sec |
| 1.1 Gpixels/sec | Horizon Detection | 5.03 Gpixels/sec |
| 1.82 Gpixels/sec | Edge Detection | 6.5 Gpixels/sec |
| 1.2 Gpixels/sec | Gaussian Blur | 7.05 Gpixels/sec |
| 0.37 Gpixels/sec | Feature Matching | 1.28 Gpixels/sec |
| 96 Gpixels/sec | Stereo Matching | 576.8 Gpixels/sec |
| 3825.9 FPS | Particle Physics | 15732.9 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 19452 |
| — | G2D Mark | 919 |
| — | DirectX 11 | 167 FPS |
| — | DirectX 12 | 75 FPS |
| — | GPU Compute | 8204 Ops/s |
| 6206 | GB6 ML Single Precision | 16090 |
| 6893 | GB6 ML Half Precision | 26121 |
| 6515 | GB6 ML Quantized | 10870 |
| 3242 | Image Classification (SP) | 6780 |
| 4890 | Image Segmentation (HP) | 15029 |
| 7149 | Image Super Resolution (Q) | 15410 |
| 10340 | Face Detection (HP) | 30849 |
| 21268 | Pose Estimation (Q) | 71857 |
| 2268 | Text Classification (SP) | 1983 |
| 1122 | Machine Translation (HP) | 3417 |
| 3007 | Object Detection (SP) | 8214 |
| 13941 | Depth Estimation (Q) | 26815 |
| 35943 | Style Transfer (SP) | 148497 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 45 GPixel/s | Pixel Fill Rate | 116 GPixel/s |
| 89 GTexel/s | Texture Fill Rate | 348 GTexel/s |
| 2.9 TFLOPS | FLOPS (FP32) | 11.2 TFLOPS |
| System Shared | Memory Type | GDDR6 |
| — | Memory Size | 8 GB |
| 6400 MHz | Memory Clock | 1937 MHz |
| 12800 Mbps | Effective Memory Speed | 15500 Mbps |
| 128-bit | Bus | 256-bit |
| No | ECC | No |
| 102.4 GB/s | Memory Bandwidth | 495.9 GB/s |
| Custom | Interface | PCIe 3.0 x16 |
| 15 W | TGP | 250 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 12 nm |
| — | Die Size | 545 mm² |
| 11.6 billion | Transistor Count | 13 billion |
| — | Transistor Density | 23.85 MTr/mm² |
| — | DirectX | 12 |
| — | Vulkan | 1.3 |
| — | OpenGL | 4.6 |
| — | OpenCL | 3 |
| — | CUDA | 7.5 |
| No | Ray Tracing | Yes |
| No | DLSS | DLSS 2 |
| — | DisplayPort | 1.4a |
| Apple | Vendor | Nvidia |
| Integrated | Build | Discrete |
| June 6, 2022 | Released | July 23, 2019 |
| Laptop | Case | Desktop |
| Professional | Purpose | Gaming |
| Mid-range | Segment | High-end |
| Apple M GPU | Architecture | Turing |
| — | GPU Codename | TU104 |
| -Apple M2or above | Recommended CPU | -Intel Core i9 9900Kor above |
| 2312 | Steel Nomad Lite Score | 11569 |
| — | Time Spy | 11621 |
| 7310 | Solar Bay | 52083 |
| — | Port Royal | 7001 |
| — | Fire Strike | 28788 |
| 5675 | Wild Life Extreme | 23792 |
| — | Night Raid | 108111 |
| 500 MHz | Base Clock | 1650 MHz |
| 1398 MHz | Boost Clock | 1815 MHz |
| 1024 | Shading Units | 3072 |
| 64 | Texture Mapping Units (TMUs) | 192 |
| 32 | Render Output Units (ROPs) | 64 |
| 128 | Compute Units (Pipelines) | 48 |
| — | Tensor Cores | 384 |
| — | Ray-tracing Cores | 48 |
| — | L1 Cache | 64KB per cluster |
| — | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |