Quick/Comparisons
The GeForce RTX 5090's exceptional gaming performance, with a perfect score, makes it the clear winner in this category. Its Blackwell 2.0 architecture and 32GB GDDR7 memory enable lightning-fast frame rates.
The GeForce RTX 5090's strong workstation performance, with a score of 93, gives it the edge in content creation tasks. Its Blackwell 2.0 architecture and 32GB GDDR7 memory provide ample resources for demanding workflows.
The GeForce RTX 5090's capable AI/ML compute, with a score of 95, makes it the winner in this category. Its Blackwell 2.0 architecture and 32GB GDDR7 memory enable efficient processing of complex AI workloads.
100 | Gaming | 34 |
93 | Workstation | 36 |
95 | AI/ML | 30 |
49 | Energy Efficiency | 47 |
96 | Quick Comparison Final Score | 36 |
| GB202-300 | Graphics Processor | Navi 22 |
| 21760 | Cores | 2560 |
| 680 / 176 | TMUs / ROPs | 160 / 64 |
| 15010.34 | Blender GPU | 1394.16 |
| 244 | Forza Horizon 5 | 82 |
| 323 | The Witcher 3 | 113 |
| 265 | Counter-Strike 2 | 147 |
| 202 | Far Cry 6 | 102 |
| 207 | Hogwarts Legacy | 67 |
| 327 | Call of Duty | 102 |
| 179 | Ghost of Tsushima | 69 |
| 229 | Cyberpunk 2077 | 72 |
| 299 | Tomb Raider | 121 |
| 253 | Average FPS | 97 |
| 294 | 1080p High | 120 |
| 253 | 1080p Ultra | 97 |
| 207 | 1440p Ultra | 69 |
| 124 | 4K Ultra | 37 |
| Low | Margin of Error | Low |
| 419309 | GB6 Compute Score | 99418 |
| 333 img/sec | Background Blur | 222.9 img/sec |
| 298.5 img/sec | Face Detection | 135.3 img/sec |
| 21.7 Gpixels/sec | Horizon Detection | 3.46 Gpixels/sec |
| 33.7 Gpixels/sec | Edge Detection | 5.14 Gpixels/sec |
| 34.5 Gpixels/sec | Gaussian Blur | 3.64 Gpixels/sec |
| 2.53 Gpixels/sec | Feature Matching | 0.92 Gpixels/sec |
| 2850 Gpixels/sec | Stereo Matching | 432.6 Gpixels/sec |
| 59146.9 FPS | Particle Physics | 13677.6 FPS |
| OpenCL | API | OpenCL |
| 38955 | G3D Mark Score | 13236 |
| 1413 | G2D Mark | 532 |
| 341 FPS | DirectX 11 | 127 FPS |
| 178 FPS | DirectX 12 | 65 FPS |
| 24632 Ops/s | GPU Compute | 5050 Ops/s |
| 114685 | Face Detection (HP) | 23070 |
| 344729 | Pose Estimation (Q) | 93148 |
| 4464 | Text Classification (SP) | 2165 |
| 8184 | Machine Translation (HP) | 3022 |
| 27802 | Object Detection (SP) | 8375 |
| 85572 | Depth Estimation (Q) | 31430 |
| 687080 | Style Transfer (SP) | 178071 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 53861 | GB6 ML Single Precision | 17087 |
| 81080 | GB6 ML Half Precision | 18189 |
| 38938 | GB6 ML Quantized | 13247 |
| 18948 | Image Classification (SP) | 7861 |
| 55109 | Image Segmentation (HP) | 7803 |
| 58478 | Image Super Resolution (Q) | 21982 |
| 424 GPixel/s | Pixel Fill Rate | 147 GPixel/s |
| 1637 GTexel/s | Texture Fill Rate | 368 GTexel/s |
| 104.8 TFLOPS | FLOPS (FP32) | 11.8 TFLOPS |
| GDDR7 | Memory Type | GDDR6 |
| 32 GB | Memory Size | 12 GB |
| 1750 MHz | Memory Clock | 1750 MHz |
| 28000 Mbps | Effective Memory Speed | 14000 Mbps |
| 512-bit | Bus | 192-bit |
| No | ECC | No |
| 1792 GB/s | Memory Bandwidth | 336 GB/s |
| PCIe 5.0 x16 | Interface | PCIe 4.0 x16 |
| 575 W | TGP | 145 W |
| TSMC | Manufacturing | TSMC |
| 4 nm | Fabrication Process | 7 nm |
| 750 mm² | Die Size | 335 mm² |
| 92.2 billion | Transistor Count | 17.2 billion |
| 122.93 MTr/mm² | Transistor Density | 51.34 MTr/mm² |
| 12.2 | DirectX | 12 |
| 1.4 | Vulkan | 1.2 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.1 |
| 12 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 4 | DLSS | No |
| 2.1b | DisplayPort | 1.4 |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| January 7, 2025 | Released | June 1, 2021 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Gaming |
| High-end | Segment | High-end |
| Blackwell 2.0 | Architecture | RDNA 2.0 |
| GB202-300 | GPU Codename | Navi 22 |
| -Intel Core Ultra 9 285Kor above | Recommended CPU | -AMD Ryzen 9 5900HXor above |
| 52638 | Steel Nomad Lite Score | 10120 |
| 46941 | Time Spy | 11506 |
| 235224 | Solar Bay | 44266 |
| 38113 | Port Royal | 5558 |
| 87433 | Fire Strike | 33464 |
| 108736 | Wild Life Extreme | 18898 |
| 207058 | Night Raid | 77742 |
| 2017 MHz | Base Clock | 2116 MHz |
| 2407 MHz | Boost Clock | 2300 MHz |
| 21760 | Shading Units | 2560 |
| 680 | Texture Mapping Units (TMUs) | 160 |
| 176 | Render Output Units (ROPs) | 64 |
| 170 | Compute Units (Pipelines) | 40 |
| 680 | Tensor Cores | — |
| 170 | Ray-tracing Cores | 40 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 96MB shared | L2 Cache | 3MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |
Compare with something else?
Compare GeForce RTX 5090 with another →