Quick/Comparisons
14 | Gaming | 50 |
23 | Workstation | 47 |
10 | AI/ML |
Compare with something else?
Compare Apple M3 GPU with another →42 |
95 | Energy Efficiency | 48 |
30 | Quick Comparison Final Score | 48 |
| — | Graphics Processor | GB206-300 |
| 1280 | Cores | 4608 |
| 80 / 40 | TMUs / ROPs | 144 / 48 |
| 155.27 | Blender GPU | 4313.42 |
| — | Forza Horizon 5 | 104 |
| — | The Witcher 3 | 147 |
| — | Counter-Strike 2 | 205 |
| — | Far Cry 6 | 125 |
| — | Hogwarts Legacy | 90 |
| — | Call of Duty | 131 |
| — | Ghost of Tsushima | 81 |
| — | Cyberpunk 2077 | 106 |
| — | Tomb Raider | 171 |
| — | Average FPS | 129 |
| — | 1080p High | 155 |
| — | 1080p Ultra | 129 |
| — | 1440p Ultra | 99 |
| — | 4K Ultra | 57 |
| — | Margin of Error | Low |
| 30027 | GB6 Compute Score | 141028 |
| 60.3 img/sec | Background Blur | 197.7 img/sec |
| 40.8 img/sec | Face Detection | 131.5 img/sec |
| 1.21 Gpixels/sec | Horizon Detection | 5.28 Gpixels/sec |
| 1.64 Gpixels/sec | Edge Detection | 6.79 Gpixels/sec |
| 1.2 Gpixels/sec | Gaussian Blur | 7.74 Gpixels/sec |
| 0.37 Gpixels/sec | Feature Matching | 1.53 Gpixels/sec |
| 102.6 Gpixels/sec | Stereo Matching | 726.7 Gpixels/sec |
| 3293.3 FPS | Particle Physics | 21833.6 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 22659 |
| — | G2D Mark | 1212 |
| — | DirectX 11 | 219 FPS |
| — | DirectX 12 | 89 FPS |
| — | GPU Compute | 11327 Ops/s |
| 4120 | GB6 ML Single Precision | 23734 |
| 6992 | GB6 ML Half Precision | 43218 |
| 5638 | GB6 ML Quantized | 18191 |
| 1579 | Image Classification (SP) | 9511 |
| 2944 | Image Segmentation (HP) | 29012 |
| 4425 | Image Super Resolution (Q) | 25621 |
| 6661 | Face Detection (HP) | 52204 |
| 15020 | Pose Estimation (Q) | 122234 |
| 2984 | Text Classification (SP) | 3209 |
| 5876 | Machine Translation (HP) | 6074 |
| 1826 | Object Detection (SP) | 12095 |
| 9339 | Depth Estimation (Q) | 45666 |
| 22291 | Style Transfer (SP) | 182591 |
| Core ML | Framework | ONNX |
| CPU | Backend | DirectML |
| 55 GPixel/s | Pixel Fill Rate | 123 GPixel/s |
| 110 GTexel/s | Texture Fill Rate | 370 GTexel/s |
| 3.5 TFLOPS | FLOPS (FP32) | 23.7 TFLOPS |
| System Shared | Memory Type | GDDR7 |
| — | Memory Size | 8 GB |
| 6400 MHz | Memory Clock | 1750 MHz |
| 12800 Mbps | Effective Memory Speed | 28000 Mbps |
| 128-bit | Bus | 128-bit |
| No | ECC | No |
| 102.4 GB/s | Memory Bandwidth | 448 GB/s |
| Custom | Interface | PCIe 5.0 x8 |
| 15 W | TGP | 180 W |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 5 nm |
| — | Die Size | 181 mm² |
| 13.8 billion | Transistor Count | 21.9 billion |
| — | Transistor Density | 120.99 MTr/mm² |
| — | DirectX | 12.2 |
| — | Vulkan | 1.4 |
| — | OpenGL | 4.6 |
| — | OpenCL | 3 |
| — | CUDA | 12 |
| Yes | Ray Tracing | Yes |
| No | DLSS | DLSS 4 |
| — | DisplayPort | 2.1b |
| Apple | Vendor | Nvidia |
| Integrated | Build | Discrete |
| October 31, 2023 | Released | April 15, 2025 |
| Laptop | Case | Desktop |
| Professional | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Apple M GPU | Architecture | Blackwell 2.0 |
| — | GPU Codename | GB206-300 |
| -Apple M3or above | Recommended CPU | -Intel Core Ultra 5 245Kor above |
| 3478 | Steel Nomad Lite Score | 16172 |
| — | Time Spy | 15835 |
| 13529 | Solar Bay | 75721 |
| — | Port Royal | 10077 |
| — | Fire Strike | 41970 |
| 8229 | Wild Life Extreme | 30331 |
| — | Night Raid | 140626 |
| 500 MHz | Base Clock | 2407 MHz |
| 1380 MHz | Boost Clock | 2572 MHz |
| 1280 | Shading Units | 4608 |
| 80 | Texture Mapping Units (TMUs) | 144 |
| 40 | Render Output Units (ROPs) | 48 |
| 160 | Compute Units (Pipelines) | 36 |
| — | Tensor Cores | 144 |
| — | Ray-tracing Cores | 36 |
| — | L1 Cache | 128KB per cluster |
| — | L2 Cache | 32MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |