Quick/Comparisons
The GeForce RTX 5090 dominates with a perfect 100 score, thanks to its exceptional gaming performance. Its superior architecture and 32GB GDDR7 memory give it a significant edge over the GeForce RTX 4090 Laptop.
The GeForce RTX 5090 takes the lead in content creation, with a 93 score that reflects its strong workstation performance and robust feature set. Although the GeForce RTX 4090 Laptop is close, it falls just short of the GeForce RTX 5090's capabilities.
The GeForce RTX 5090 excels in AI and machine learning tasks, with a 95 score that highlights its capable compute performance and advanced architecture. The GeForce RTX 4090 Laptop trails behind, but still offers respectable AI and ML capabilities.
63 | Gaming | 100 |
63 | Workstation | 93 |
55 | AI/ML | 95 |
45 | Energy Efficiency | 49 |
59 | Quick Comparison Final Score | 96 |
| AD103 (GN21-X11) | Graphics Processor | GB202-300 |
| 9728 | Cores | 21760 |
| 304 / 112 | TMUs / ROPs | 680 / 176 |
| 7059.06 | Blender GPU | 15010.34 |
| 131 | Forza Horizon 5 | 244 |
| 182 | The Witcher 3 | 323 |
| 208 | Counter-Strike 2 | 265 |
| 147 | Far Cry 6 | 202 |
| 108 | Hogwarts Legacy | 207 |
| 164 | Call of Duty | 327 |
| 100 | Ghost of Tsushima | 179 |
| 137 | Cyberpunk 2077 | 229 |
| 231 | Tomb Raider | 299 |
| 156 | Average FPS | 253 |
| 187 | 1080p High | 294 |
| 156 | 1080p Ultra | 253 |
| 123 | 1440p Ultra | 207 |
| 72 | 4K Ultra | 124 |
| Low | Margin of Error | Low |
| 203263 | GB6 Compute Score | 419309 |
| 264 img/sec | Background Blur | 333 img/sec |
| 134.3 img/sec | Face Detection | 298.5 img/sec |
| 9.27 Gpixels/sec | Horizon Detection | 21.7 Gpixels/sec |
| 12.7 Gpixels/sec | Edge Detection | 33.7 Gpixels/sec |
| 13.2 Gpixels/sec | Gaussian Blur | 34.5 Gpixels/sec |
| 1.92 Gpixels/sec | Feature Matching | 2.53 Gpixels/sec |
| 1060 Gpixels/sec | Stereo Matching | 2850 Gpixels/sec |
| 29225.1 FPS | Particle Physics | 59146.9 FPS |
| OpenCL | API | OpenCL |
| 27060 | G3D Mark Score | 38955 |
| 988 | G2D Mark | 1413 |
| 258 FPS | DirectX 11 | 341 FPS |
| 106 FPS | DirectX 12 | 178 FPS |
| 12206 Ops/s | GPU Compute | 24632 Ops/s |
| 29362 | GB6 ML Single Precision | 53861 |
| 42611 | GB6 ML Half Precision | 81080 |
| 22212 | GB6 ML Quantized | 38938 |
| 11346 | Image Classification (SP) | 18948 |
| 24625 | Image Segmentation (HP) | 55109 |
| 36874 | Image Super Resolution (Q) | 58478 |
| 50088 | Face Detection (HP) | 114685 |
| 174844 | Pose Estimation (Q) | 344729 |
| 3226 | Text Classification (SP) | 4464 |
| 4905 | Machine Translation (HP) | 8184 |
| 14667 | Object Detection (SP) | 27802 |
| 53868 | Depth Estimation (Q) | 85572 |
| 340056 | Style Transfer (SP) | 687080 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 163-228 GPixel/s | Pixel Fill Rate | 424 GPixel/s |
| 442-620 GTexel/s | Texture Fill Rate | 1637 GTexel/s |
| 28.3-39.7 TFLOPS | FLOPS (FP32) | 104.8 TFLOPS |
| GDDR6 | Memory Type | GDDR7 |
| 16 GB | Memory Size | 32 GB |
| 2000 MHz | Memory Clock | 1750 MHz |
| 16000 Mbps | Effective Memory Speed | 28000 Mbps |
| 256-bit | Bus | 512-bit |
| No | ECC | No |
| 512 GB/s | Memory Bandwidth | 1792 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 80-175 W (configurable) | TGP | 575 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 4 nm |
| 379 mm² | Die Size | 750 mm² |
| 45.9 billion | Transistor Count | 92.2 billion |
| 121.11 MTr/mm² | Transistor Density | 122.93 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 | January 7, 2025 |
| Laptop | Case | Desktop |
| Gaming | Purpose | Gaming |
| High-end | Segment | High-end |
| Ada Lovelace | Architecture | Blackwell 2.0 |
| AD103 (GN21-X11) | GPU Codename | GB202-300 |
| -Intel Core Ultra 9 185Hor above | Recommended CPU | -Intel Core Ultra 9 285Kor above |
| 22501 | Steel Nomad Lite Score | 52638 |
| 21401 | Time Spy | 46941 |
| 96130 | Solar Bay | 235224 |
| 13733 | Port Royal | 38113 |
| 45218 | Fire Strike | 87433 |
| 42491 | Wild Life Extreme | 108736 |
| 147892 | Night Raid | 207058 |
| 930-1620 MHz (configurable) | Base Clock | 2017 MHz |
| 1455-2040 MHz (configurable) | Boost Clock | 2407 MHz |
| 9728 | Shading Units | 21760 |
| 304 | Texture Mapping Units (TMUs) | 680 |
| 112 | Render Output Units (ROPs) | 176 |
| 76 | Compute Units (Pipelines) | 170 |
| 304 | Tensor Cores | 680 |
| 76 | Ray-tracing Cores | 170 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 64MB shared | L2 Cache | 96MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |
Compare with something else?
Compare GeForce RTX 4090 with another →