Quick/Comparisons
The GeForce RTX 5090 takes the lead in gaming performance, thanks to its 100 score, outpacing the Nvidia RTX 6000 Ada's 80.
Both the GeForce RTX 5090 and Nvidia RTX 6000 Ada deliver strong workstation performance, with the Nvidia RTX 6000 Ada edging out the GeForce RTX 5090 by a single point.
The GeForce RTX 5090 excels in AI/ML compute, boasting a 95 score, significantly outperforming the Nvidia RTX 6000 Ada's 75.
100 | Gaming | 80 |
93 | Workstation | 94 |
95 | AI/ML |
Compare with something else?
Compare GeForce RTX 5090 with another →75 |
49 | Energy Efficiency | 44 |
96 | Quick Comparison Final Score | 83 |
| GB202-300 | Graphics Processor | AD102 |
| 21760 | Cores | 18176 |
| 680 / 176 | TMUs / ROPs | 568 / 192 |
| 15010.34 | Blender GPU | 11038.91 |
| 244 | Forza Horizon 5 | 165 |
| 323 | The Witcher 3 | 219 |
| 265 | Counter-Strike 2 | 221 |
| 202 | Far Cry 6 | 164 |
| 207 | Hogwarts Legacy | 133 |
| 327 | Call of Duty | 216 |
| 179 | Ghost of Tsushima | 123 |
| 229 | Cyberpunk 2077 | 172 |
| 299 | Tomb Raider | 267 |
| 253 | Average FPS | 187 |
| 294 | 1080p High | 219 |
| 253 | 1080p Ultra | 187 |
| 207 | 1440p Ultra | 148 |
| 124 | 4K Ultra | 89 |
| Low | Margin of Error | High |
| 419309 | GB6 Compute Score | 389848 |
| 333 img/sec | Background Blur | 618.3 img/sec |
| 298.5 img/sec | Face Detection | 399.9 img/sec |
| 21.7 Gpixels/sec | Horizon Detection | 15.1 Gpixels/sec |
| 33.7 Gpixels/sec | Edge Detection | 22.1 Gpixels/sec |
| 34.5 Gpixels/sec | Gaussian Blur | 26.3 Gpixels/sec |
| 2.53 Gpixels/sec | Feature Matching | 2.33 Gpixels/sec |
| 2850 Gpixels/sec | Stereo Matching | 2210 Gpixels/sec |
| 59146.9 FPS | Particle Physics | 53494.5 FPS |
| OpenCL | API | OpenCL |
| 38955 | G3D Mark Score | 28613 |
| 1413 | G2D Mark | 1005 |
| 341 FPS | DirectX 11 | 284 FPS |
| 178 FPS | DirectX 12 | 102 FPS |
| 24632 Ops/s | GPU Compute | 22439 Ops/s |
| 114685 | Face Detection (HP) | 69004 |
| 344729 | Pose Estimation (Q) | 230965 |
| 4464 | Text Classification (SP) | 3491 |
| 8184 | Machine Translation (HP) | 5545 |
| 27802 | Object Detection (SP) | 18668 |
| 85572 | Depth Estimation (Q) | 68370 |
| 687080 | Style Transfer (SP) | 520389 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 53861 | GB6 ML Single Precision | 37270 |
| 81080 | GB6 ML Half Precision | 53469 |
| 38938 | GB6 ML Quantized | 27850 |
| 18948 | Image Classification (SP) | 13367 |
| 55109 | Image Segmentation (HP) | 32477 |
| 58478 | Image Super Resolution (Q) | 42056 |
| 424 GPixel/s | Pixel Fill Rate | 481 GPixel/s |
| 1637 GTexel/s | Texture Fill Rate | 1423 GTexel/s |
| 104.8 TFLOPS | FLOPS (FP32) | 91.1 TFLOPS |
| GDDR7 | Memory Type | GDDR6 |
| 32 GB | Memory Size | 48 GB |
| 1750 MHz | Memory Clock | 2500 MHz |
| 28000 Mbps | Effective Memory Speed | 20000 Mbps |
| 512-bit | Bus | 384-bit |
| No | ECC | No |
| 1792 GB/s | Memory Bandwidth | 960 GB/s |
| PCIe 5.0 x16 | Interface | PCIe 4.0 x16 |
| 575 W | TGP | 300 W |
| TSMC | Manufacturing | TSMC |
| 4 nm | Fabrication Process | 5 nm |
| 750 mm² | Die Size | 609 mm² |
| 92.2 billion | Transistor Count | 76 billion |
| 122.93 MTr/mm² | Transistor Density | 124.79 MTr/mm² |
| 12.2 | DirectX | 12 |
| 1.4 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 12 | CUDA | 8.9 |
| Yes | Ray Tracing | Yes |
| DLSS 4 | DLSS | DLSS 3 |
| 2.1b | DisplayPort | 1.4a |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| January 7, 2025 | Released | December 3, 2022 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Professional |
| High-end | Segment | High-end |
| Blackwell 2.0 | Architecture | Ada Lovelace |
| GB202-300 | GPU Codename | AD102 |
| -Intel Core Ultra 9 285Kor above | Recommended CPU | -Intel Core i9 14900Kor above |
| 52638 | Steel Nomad Lite Score | 32284 |
| 46941 | Time Spy | 26686 |
| 235224 | Solar Bay | 145429 |
| 38113 | Port Royal | 19557 |
| 87433 | Fire Strike | 62803 |
| 108736 | Wild Life Extreme | 53640 |
| 207058 | Night Raid | 148225 |
| 2017 MHz | Base Clock | 915 MHz |
| 2407 MHz | Boost Clock | 2505 MHz |
| 21760 | Shading Units | 18176 |
| 680 | Texture Mapping Units (TMUs) | 568 |
| 176 | Render Output Units (ROPs) | 192 |
| 170 | Compute Units (Pipelines) | 142 |
| 680 | Tensor Cores | 568 |
| 170 | Ray-tracing Cores | 142 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 96MB shared | L2 Cache | 96MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |