Quick/Comparisons
100 | Gaming | 4 |
93 | Workstation | 10 |
95 | AI/ML |
Compare with something else?
Compare GeForce RTX 5090 with another →2 |
49 | Energy Efficiency | 71 |
96 | Quick Comparison Final Score | 17 |
| GB202-300 | Graphics Processor | Picasso |
| 21760 | Cores | 192 |
| 680 / 176 | TMUs / ROPs | 12 / 4 |
| 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 | 4457 |
| 333 img/sec | Background Blur | 10.1 img/sec |
| 298.5 img/sec | Face Detection | 4.97 img/sec |
| 21.7 Gpixels/sec | Horizon Detection | 0.16 Gpixels/sec |
| 33.7 Gpixels/sec | Edge Detection | 0.33 Gpixels/sec |
| 34.5 Gpixels/sec | Gaussian Blur | 0.22 Gpixels/sec |
| 2.53 Gpixels/sec | Feature Matching | 0.04 Gpixels/sec |
| 2850 Gpixels/sec | Stereo Matching | 14.1 Gpixels/sec |
| 59146.9 FPS | Particle Physics | 500.2 FPS |
| OpenCL | API | OpenCL |
| 38955 | G3D Mark Score | 892 |
| 1413 | G2D Mark | 302 |
| 341 FPS | DirectX 11 | 7 FPS |
| 178 FPS | DirectX 12 | 4 FPS |
| 24632 Ops/s | GPU Compute | 442 Ops/s |
| 114685 | Face Detection (HP) | 1535 |
| 344729 | Pose Estimation (Q) | 2770 |
| 4464 | Text Classification (SP) | 430 |
| 8184 | Machine Translation (HP) | 706 |
| 27802 | Object Detection (SP) | 532 |
| 85572 | Depth Estimation (Q) | 1792 |
| 687080 | Style Transfer (SP) | 5517 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 53861 | GB6 ML Single Precision | 1122 |
| 81080 | GB6 ML Half Precision | 1342 |
| 38938 | GB6 ML Quantized | 815 |
| 18948 | Image Classification (SP) | 476 |
| 55109 | Image Segmentation (HP) | 498 |
| 58478 | Image Super Resolution (Q) | 982 |
| 424 GPixel/s | Pixel Fill Rate | 4-5 GPixel/s |
| 1637 GTexel/s | Texture Fill Rate | 12-14 GTexel/s |
| 104.8 TFLOPS | FLOPS (FP32) | 0.4-0.5 TFLOPS |
| GDDR7 | Memory Type | System Shared |
| 32 GB | Memory Size | System RAM Dependent |
| 1750 MHz | Memory Clock | System RAM Dependent |
| 28000 Mbps | Effective Memory Speed | System RAM Dependent |
| 512-bit | Bus | — |
| No | ECC | No |
| 1792 GB/s | Memory Bandwidth | — |
| PCIe 5.0 x16 | Interface | IGP |
| 575 W | TGP | 15 W |
| TSMC | Manufacturing | GlobalFoundries |
| 4 nm | Fabrication Process | 14 nm |
| 750 mm² | Die Size | 210 mm² |
| 92.2 billion | Transistor Count | 4.94 billion |
| 122.93 MTr/mm² | Transistor Density | 23.52 MTr/mm² |
| 12.2 | DirectX | 12 |
| 1.4 | Vulkan | 1.2.131 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2 |
| 12 | CUDA | — |
| Yes | Ray Tracing | No |
| DLSS 4 | DLSS | No |
| 2.1b | DisplayPort | — |
| Nvidia | Vendor | Amd |
| Discrete | Build | Integrated |
| January 7, 2025 | Released | January 1, 2019 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Generic |
| High-end | Segment | Entry-level |
| Blackwell 2.0 | Architecture | GCN 5.0 |
| GB202-300 | GPU Codename | Picasso |
| -Intel Core Ultra 9 285Kor above | Recommended CPU | -AMD Ryzen 3 3200Uor above |
| 52638 | Steel Nomad Lite Score | 390 |
| 46941 | Time Spy | 412 |
| 235224 | Solar Bay | — |
| 38113 | Port Royal | — |
| 87433 | Fire Strike | 1448 |
| 108736 | Wild Life Extreme | 867 |
| 207058 | Night Raid | 6135 |
| 2017 MHz | Base Clock | 300 MHz |
| 2407 MHz | Boost Clock | 1000-1200 MHz (configurable) |
| 21760 | Shading Units | 192 |
| 680 | Texture Mapping Units (TMUs) | 12 |
| 176 | Render Output Units (ROPs) | 4 |
| 170 | Compute Units (Pipelines) | 3 |
| 680 | Tensor Cores | No |
| 170 | Ray-tracing Cores | No |
| 128KB per cluster | L1 Cache | — |
| 96MB shared | L2 Cache | — |
| 2 IPC | Instructions Per Cycle | 2 IPC |