Quick/Comparisons
100 | Gaming | 16 |
93 | Workstation | 10 |
95 | AI/ML |
Compare with something else?
Compare GeForce RTX 5090 with another →7 |
49 | Energy Efficiency | 94 |
96 | Quick Comparison Final Score | 26 |
| GB202-300 | Graphics Processor | — |
| 21760 | Cores | 512 |
| 680 / 176 | TMUs / ROPs | 32 / 16 |
| 15010.34 | Blender GPU | — |
| 244 | Forza Horizon 5 | — |
| 323 | The Witcher 3 | — |
| 265 | Counter-Strike 2 | — |
| 202 | Far Cry 6 | — |
| 207 | Hogwarts Legacy | — |
| 327 | Call of Duty | — |
| 179 | Ghost of Tsushima | — |
| 229 | Cyberpunk 2077 | — |
| 299 | Tomb Raider | — |
| 253 | Average FPS | — |
| 294 | 1080p High | — |
| 253 | 1080p Ultra | — |
| 207 | 1440p Ultra | — |
| 124 | 4K Ultra | — |
| Low | Margin of Error | — |
| 419309 | GB6 Compute Score | 16321 |
| 333 img/sec | Background Blur | 37.6 img/sec |
| 298.5 img/sec | Face Detection | 17.2 img/sec |
| 21.7 Gpixels/sec | Horizon Detection | 0.3 Gpixels/sec |
| 33.7 Gpixels/sec | Edge Detection | 0.63 Gpixels/sec |
| 34.5 Gpixels/sec | Gaussian Blur | 0.94 Gpixels/sec |
| 2.53 Gpixels/sec | Feature Matching | 0.22 Gpixels/sec |
| 2850 Gpixels/sec | Stereo Matching | 60 Gpixels/sec |
| 59146.9 FPS | Particle Physics | 3695.9 FPS |
| OpenCL | API | OpenCL |
| 38955 | G3D Mark Score | — |
| 1413 | G2D Mark | — |
| 341 FPS | DirectX 11 | — |
| 178 FPS | DirectX 12 | — |
| 24632 Ops/s | GPU Compute | — |
| 114685 | Face Detection (HP) | 6068 |
| 344729 | Pose Estimation (Q) | 8220 |
| 4464 | Text Classification (SP) | 609 |
| 8184 | Machine Translation (HP) | 1146 |
| 27802 | Object Detection (SP) | 2766 |
| 85572 | Depth Estimation (Q) | 6405 |
| 687080 | Style Transfer (SP) | 26796 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 53861 | GB6 ML Single Precision | 3770 |
| 81080 | GB6 ML Half Precision | 5146 |
| 38938 | GB6 ML Quantized | 2824 |
| 18948 | Image Classification (SP) | 2902 |
| 55109 | Image Segmentation (HP) | 5473 |
| 58478 | Image Super Resolution (Q) | 2967 |
| 424 GPixel/s | Pixel Fill Rate | 28-32 GPixel/s |
| 1637 GTexel/s | Texture Fill Rate | 56-64 GTexel/s |
| 104.8 TFLOPS | FLOPS (FP32) | 1.8-2 TFLOPS |
| GDDR7 | Memory Type | System Shared |
| 32 GB | Memory Size | — |
| 1750 MHz | Memory Clock | — |
| 28000 Mbps | Effective Memory Speed | — |
| 512-bit | Bus | — |
| No | ECC | — |
| 1792 GB/s | Memory Bandwidth | — |
| PCIe 5.0 x16 | Interface | Ring Bus |
| 575 W | TGP | 14 W |
| TSMC | Manufacturing | TSMC |
| 4 nm | Fabrication Process | 5 nm |
| 750 mm² | Die Size | — |
| 92.2 billion | Transistor Count | — |
| 122.93 MTr/mm² | Transistor Density | — |
| 12.2 | DirectX | 12 |
| 1.4 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 12 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 4 | DLSS | No |
| 2.1b | DisplayPort | 2.1 |
| Nvidia | Vendor | Intel |
| Discrete | Build | Integrated |
| January 7, 2025 | Released | December 14, 2023 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Generic |
| High-end | Segment | Mid-range |
| Blackwell 2.0 | Architecture | Xe LPG |
| GB202-300 | GPU Codename | — |
| -Intel Core Ultra 9 285Kor above | Recommended CPU | -Intel Core Ultra 5 125Uor above |
| 52638 | Steel Nomad Lite Score | — |
| 46941 | Time Spy | — |
| 235224 | Solar Bay | — |
| 38113 | Port Royal | — |
| 87433 | Fire Strike | — |
| 108736 | Wild Life Extreme | — |
| 207058 | Night Raid | — |
| 2017 MHz | Base Clock | 300 MHz |
| 2407 MHz | Boost Clock | 1750-2000 MHz (configurable) |
| 21760 | Shading Units | 512 |
| 680 | Texture Mapping Units (TMUs) | 32 |
| 176 | Render Output Units (ROPs) | 16 |
| 170 | Compute Units (Pipelines) | 64 |
| 680 | Tensor Cores | — |
| 170 | Ray-tracing Cores | 4 |
| 128KB per cluster | L1 Cache | 768KB shared |
| 96MB shared | L2 Cache | — |
| 2 IPC | Instructions Per Cycle | 2 IPC |