Quick/Comparisons
51 | Gaming | 80 |
57 | Workstation | 94 |
42 | AI/ML |
Compare with something else?
Compare Nvidia RTX 4000 with another →75 |
45 | Energy Efficiency | 44 |
51 | Quick Comparison Final Score | 83 |
| AD104 | Graphics Processor | AD102 |
| 7424 | Cores | 18176 |
| 232 / 80 | TMUs / ROPs | 568 / 192 |
| 5009.03 | Blender GPU | 11038.91 |
| 107 | Forza Horizon 5 | 165 |
| 152 | The Witcher 3 | 219 |
| 200 | Counter-Strike 2 | 221 |
| 126 | Far Cry 6 | 164 |
| 89 | Hogwarts Legacy | 133 |
| 132 | Call of Duty | 216 |
| 82 | Ghost of Tsushima | 123 |
| 109 | Cyberpunk 2077 | 172 |
| 175 | Tomb Raider | 267 |
| 130 | Average FPS | 187 |
| 157 | 1080p High | 219 |
| 130 | 1080p Ultra | 187 |
| 99 | 1440p Ultra | 148 |
| 58 | 4K Ultra | 89 |
| High | Margin of Error | High |
| 139462 | GB6 Compute Score | 389848 |
| 187.1 img/sec | Background Blur | 618.3 img/sec |
| 139.5 img/sec | Face Detection | 399.9 img/sec |
| 5.71 Gpixels/sec | Horizon Detection | 15.1 Gpixels/sec |
| 8.53 Gpixels/sec | Edge Detection | 22.1 Gpixels/sec |
| 8.38 Gpixels/sec | Gaussian Blur | 26.3 Gpixels/sec |
| 1.55 Gpixels/sec | Feature Matching | 2.33 Gpixels/sec |
| 577.7 Gpixels/sec | Stereo Matching | 2210 Gpixels/sec |
| 16849 FPS | Particle Physics | 53494.5 FPS |
| OpenCL | API | OpenCL |
| 22027 | G3D Mark Score | 28613 |
| 721 | G2D Mark | 1005 |
| 215 FPS | DirectX 11 | 284 FPS |
| 96 FPS | DirectX 12 | 102 FPS |
| 8635 Ops/s | GPU Compute | 22439 Ops/s |
| 22488 | GB6 ML Single Precision | 37270 |
| 35441 | GB6 ML Half Precision | 53469 |
| 16689 | GB6 ML Quantized | 27850 |
| 8658 | Image Classification (SP) | 13367 |
| 17759 | Image Segmentation (HP) | 32477 |
| 28281 | Image Super Resolution (Q) | 42056 |
| 37971 | Face Detection (HP) | 69004 |
| 113009 | Pose Estimation (Q) | 230965 |
| 3535 | Text Classification (SP) | 3491 |
| 4038 | Machine Translation (HP) | 5545 |
| 11288 | Object Detection (SP) | 18668 |
| 38963 | Depth Estimation (Q) | 68370 |
| 236141 | Style Transfer (SP) | 520389 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 117-181 GPixel/s | Pixel Fill Rate | 481 GPixel/s |
| 340-524 GTexel/s | Texture Fill Rate | 1423 GTexel/s |
| 21.8-33.6 TFLOPS | FLOPS (FP32) | 91.1 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 12 GB | Memory Size | 48 GB |
| 2250 MHz | Memory Clock | 2500 MHz |
| 18000 Mbps | Effective Memory Speed | 20000 Mbps |
| 192-bit | Bus | 384-bit |
| Yes | ECC | No |
| 432 GB/s | Memory Bandwidth | 960 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 60-175 W (configurable) | TGP | 300 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 5 nm |
| 294 mm² | Die Size | 609 mm² |
| 35.8 billion | Transistor Count | 76 billion |
| 121.77 MTr/mm² | Transistor Density | 124.79 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.9 | CUDA | 8.9 |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | DLSS 3 |
| 1.4a | DisplayPort | 1.4a |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| March 21, 2023 | Released | December 3, 2022 |
| Laptop | Case | Desktop |
| Professional | Purpose | Professional |
| High-end | Segment | High-end |
| Ada Lovelace | Architecture | Ada Lovelace |
| AD104 | GPU Codename | AD102 |
| -Intel Core Ultra 9 185Hor above | Recommended CPU | -Intel Core i9 14900Kor above |
| 16926 | Steel Nomad Lite Score | 32284 |
| 15516 | Time Spy | 26686 |
| 68617 | Solar Bay | 145429 |
| 10118 | Port Royal | 19557 |
| 38840 | Fire Strike | 62803 |
| — | Wild Life Extreme | 53640 |
| 123923 | Night Raid | 148225 |
| — | Base Clock | 915 MHz |
| 1465-2260 MHz (configurable) | Boost Clock | 2505 MHz |
| 7424 | Shading Units | 18176 |
| 232 | Texture Mapping Units (TMUs) | 568 |
| 80 | Render Output Units (ROPs) | 192 |
| 58 | Compute Units (Pipelines) | 142 |
| 232 | Tensor Cores | 568 |
| 58 | Ray-tracing Cores | 142 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 48MB shared | L2 Cache | 96MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |