Quick/Comparisons
15 | Gaming | 31 |
25 | Workstation | 35 |
21 | AI/ML |
Compare with something else?
Compare Apple M4 GPU with another →29 |
95 | Energy Efficiency | 67 |
32 | Quick Comparison Final Score | 38 |
| Custom | Graphics Processor | GB207 |
| 1280 | Cores | 2560 |
| 80 / 40 | TMUs / ROPs | 80 / 32 |
| 1062.09 | Blender GPU | 2952.19 |
| — | Tomb Raider | 95 |
| — | Average FPS | 79 |
| — | Forza Horizon 5 | 68 |
| — | The Witcher 3 | 98 |
| — | Counter-Strike 2 | 133 |
| — | Far Cry 6 | 81 |
| — | Hogwarts Legacy | 56 |
| — | Call of Duty | 71 |
| — | Ghost of Tsushima | 49 |
| — | Cyberpunk 2077 | 63 |
| — | 1080p High | 99 |
| — | 1080p Ultra | 79 |
| — | 1440p Ultra | 57 |
| — | 4K Ultra | 31 |
| — | Margin of Error | Low |
| 37678 | GB6 Compute Score | 91033 |
| 72.2 img/sec | Background Blur | 154.5 img/sec |
| 48.3 img/sec | Face Detection | 94.6 img/sec |
| 1.4 Gpixels/sec | Horizon Detection | 3.36 Gpixels/sec |
| 1.94 Gpixels/sec | Edge Detection | 4.21 Gpixels/sec |
| 1.5 Gpixels/sec | Gaussian Blur | 4.26 Gpixels/sec |
| 0.53 Gpixels/sec | Feature Matching | 1.07 Gpixels/sec |
| 123.3 Gpixels/sec | Stereo Matching | 414.3 Gpixels/sec |
| 4938.4 FPS | Particle Physics | 13519.1 FPS |
| OpenCL | API | OpenCL |
| — | G3D Mark Score | 14357 |
| — | G2D Mark | 760 |
| — | DirectX 11 | 128 FPS |
| — | DirectX 12 | 67 FPS |
| — | GPU Compute | 5169 Ops/s |
| 10346 | GB6 ML Single Precision | 16186 |
| 11883 | GB6 ML Half Precision | 31589 |
| 11036 | GB6 ML Quantized | 13082 |
| 4874 | Image Classification (SP) | 6724 |
| 8168 | Image Segmentation (HP) | 20455 |
| 11983 | Image Super Resolution (Q) | 17332 |
| 17036 | Face Detection (HP) | 37078 |
| 31171 | Pose Estimation (Q) | 68680 |
| 2964 | Text Classification (SP) | 2764 |
| 3410 | Machine Translation (HP) | 5076 |
| 5112 | Object Detection (SP) | 8520 |
| 22700 | Depth Estimation (Q) | 32304 |
| 60872 | Style Transfer (SP) | 110367 |
| Core ML | Framework | ONNX |
| GPU | Backend | DirectML |
| 59 GPixel/s | Pixel Fill Rate | 48-85 GPixel/s |
| 118 GTexel/s | Texture Fill Rate | 120-213 GTexel/s |
| 3.8 TFLOPS | FLOPS (FP32) | 7.7-13.6 TFLOPS |
| System Shared | Memory Type | GDDR7 |
| — | Memory Size | 8 GB |
| 7500 MHz | Memory Clock | 1750 MHz |
| 15000 Mbps | Effective Memory Speed | 24000 Mbps |
| 128-bit | Bus | 128-bit |
| No | ECC | No |
| 120 GB/s | Memory Bandwidth | 384 GB/s |
| Custom | Interface | PCIe 5.0 x16 |
| 18 W | TGP | 35-115 W (configurable) |
| TSMC | Manufacturing | TSMC |
| 3 nm | Fabrication Process | 5 nm |
| — | Die Size | 149 mm² |
| 28 billion | Transistor Count | 16.9 billion |
| — | Transistor Density | 113.42 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 |
| May 7, 2024 | Released | June 24, 2025 |
| Laptop | Case | Laptop |
| Professional | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Apple M GPU | Architecture | Blackwell 2.0 |
| Custom | GPU Codename | GB207 |
| -Apple M4 (10-Core)or above | Recommended CPU | -Intel Core Ultra 5 245HXor above |
| 4097 | Steel Nomad Lite Score | 9532 |
| — | Time Spy | 9528 |
| 16545 | Solar Bay | 42379 |
| — | Port Royal | 5774 |
| — | Fire Strike | 25853 |
| 9591 | Wild Life Extreme | 17822 |
| — | Night Raid | 90582 |
| 500 MHz | Base Clock | 1020-1500 MHz (configurable) |
| 1470 MHz | Boost Clock | 1500-2662 MHz (configurable) |
| 1280 | Shading Units | 2560 |
| 80 | Texture Mapping Units (TMUs) | 80 |
| 40 | Render Output Units (ROPs) | 32 |
| 160 | Compute Units (Pipelines) | 20 |
| — | Tensor Cores | 80 |
| — | Ray-tracing Cores | 20 |
| — | L1 Cache | 128KB per cluster |
| — | L2 Cache | 32MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |