Quick/Comparisons
57 | Gaming | 66 |
55 | Workstation | 79 |
48 | AI/ML |
Compare with something else?
Compare GeForce RTX 4080 with another →62 |
45 | Energy Efficiency | 49 |
53 | Quick Comparison Final Score | 66 |
| AD104 (GN21-X9) | Graphics Processor | GB202-300-A1 |
| 7424 | Cores | 10496 |
| 232 / 80 | TMUs / ROPs | 328 / 112 |
| 5463.87 | Blender GPU | 7401.59 |
| 120 | Forza Horizon 5 | 135 |
| 163 | The Witcher 3 | 180 |
| 209 | Counter-Strike 2 | 214 |
| 139 | Far Cry 6 | 148 |
| 98 | Hogwarts Legacy | 114 |
| 153 | Call of Duty | 175 |
| 92 | Ghost of Tsushima | 106 |
| 122 | Cyberpunk 2077 | 145 |
| 208 | Tomb Raider | 245 |
| 145 | Average FPS | 162 |
| 171 | 1080p High | 191 |
| 145 | 1080p Ultra | 162 |
| 112 | 1440p Ultra | 127 |
| 66 | 4K Ultra | 75 |
| Low | Margin of Error | High |
| 170767 | GB6 Compute Score | 246458 |
| 328.9 img/sec | Background Blur | 267.1 img/sec |
| 192.3 img/sec | Face Detection | 183.2 img/sec |
| 6.51 Gpixels/sec | Horizon Detection | 11.5 Gpixels/sec |
| 9.11 Gpixels/sec | Edge Detection | 15.8 Gpixels/sec |
| 10.1 Gpixels/sec | Gaussian Blur | 18.1 Gpixels/sec |
| 1.56 Gpixels/sec | Feature Matching | 2.01 Gpixels/sec |
| 660.7 Gpixels/sec | Stereo Matching | 1330 Gpixels/sec |
| 20765.5 FPS | Particle Physics | 35463.8 FPS |
| OpenCL | API | OpenCL |
| 24711 | G3D Mark Score | 24043 |
| 922 | G2D Mark | 843 |
| 241 FPS | DirectX 11 | 254 FPS |
| 96 FPS | DirectX 12 | 109 FPS |
| 11008 Ops/s | GPU Compute | 8247 Ops/s |
| 26637 | GB6 ML Single Precision | 31330 |
| 39211 | GB6 ML Half Precision | 53746 |
| 19995 | GB6 ML Quantized | 24267 |
| 10385 | Image Classification (SP) | 12454 |
| 25896 | Image Segmentation (HP) | 33371 |
| 27816 | Image Super Resolution (Q) | 35476 |
| 51976 | Face Detection (HP) | 67998 |
| 142459 | Pose Estimation (Q) | 168369 |
| 3405 | Text Classification (SP) | 3491 |
| 4860 | Machine Translation (HP) | 6115 |
| 13308 | Object Detection (SP) | 16115 |
| 49203 | Depth Estimation (Q) | 59893 |
| 262792 | Style Transfer (SP) | 279891 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 133-182 GPixel/s | Pixel Fill Rate | 188-202 GPixel/s |
| 386-529 GTexel/s | Texture Fill Rate | 551-593 GTexel/s |
| 24.7-33.9 TFLOPS | FLOPS (FP32) | 35.3-37.9 TFLOPS |
| GDDR6 | Memory Type | GDDR7 |
| 12 GB | Memory Size | 24 GB |
| 2250 MHz | Memory Clock | 1750 MHz |
| 18000 Mbps | Effective Memory Speed | 28000 Mbps |
| 192-bit | Bus | 256-bit |
| No | ECC | — |
| 432 GB/s | Memory Bandwidth | 896 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 60-175 W (configurable) | TGP | 95-175 W (configurable) |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 5 nm |
| 295 mm² | Die Size | 378 mm² |
| 35.8 billion | Transistor Count | 45.6 billion |
| 121.36 MTr/mm² | Transistor Density | 120.63 MTr/mm² |
| 12 | DirectX | 12.2 |
| 1.3 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.9 | CUDA | 12 |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | DLSS 4 |
| 1.4a | DisplayPort | 2.1b |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| February 8, 2023 | Released | March 18, 2025 |
| Laptop | Case | Laptop |
| Gaming | Purpose | Professional |
| High-end | Segment | High-end |
| Ada Lovelace | Architecture | Blackwell 2.0 |
| AD104 (GN21-X9) | GPU Codename | GB202-300-A1 |
| -Intel Core Ultra 9 185Hor above | Recommended CPU | -Intel Core Ultra 9 275HXor above |
| 11837 | Port Royal | 14090 |
| 19263 | Steel Nomad Lite Score | 24118 |
| 18832 | Time Spy | 21548 |
| 85404 | Solar Bay | 96355 |
| 41182 | Fire Strike | 47636 |
| 36965 | Wild Life Extreme | — |
| 147212 | Night Raid | — |
| 1290-1860 MHz (configurable) | Base Clock | 990-1680 MHz (configurable) |
| 1665-2280 MHz (configurable) | Boost Clock | 1680-1807 MHz (configurable) |
| 7424 | Shading Units | 10496 |
| 232 | Texture Mapping Units (TMUs) | 328 |
| 80 | Render Output Units (ROPs) | 112 |
| 58 | Compute Units (Pipelines) | 82 |
| 232 | Tensor Cores | 328 |
| 58 | Ray-tracing Cores | 82 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 48MB shared | L2 Cache | 64MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |