Quick/Comparisons
71 | Gaming | 22 |
73 | Workstation | 17 |
60 | AI/ML |
Compare with something else?
Compare GeForce RTX 5090 with another →| — |
50 | Energy Efficiency | 66 |
67 | Quick Comparison Final Score | 28 |
| GB203-400 | Graphics Processor | Strix Halo |
| 10496 | Cores | 1024 |
| 328 / 112 | TMUs / ROPs | 64 / 32 |
| 8001.96 | Blender GPU | 376.14 |
| 154 | Forza Horizon 5 | — |
| 198 | The Witcher 3 | — |
| 213 | Counter-Strike 2 | — |
| 155 | Far Cry 6 | — |
| 123 | Hogwarts Legacy | — |
| 187 | Call of Duty | — |
| 115 | Ghost of Tsushima | — |
| 155 | Cyberpunk 2077 | — |
| 260 | Tomb Raider | — |
| 173 | Average FPS | — |
| 202 | 1080p High | — |
| 173 | 1080p Ultra | — |
| 137 | 1440p Ultra | — |
| 80 | 4K Ultra | — |
| Low | Margin of Error | — |
| 226752 | GB6 Compute Score | 35976 |
| 266.5 img/sec | Background Blur | 106.8 img/sec |
| 196.9 img/sec | Face Detection | 43.8 img/sec |
| 10.6 Gpixels/sec | Horizon Detection | 1.12 Gpixels/sec |
| 15.2 Gpixels/sec | Edge Detection | 1.17 Gpixels/sec |
| 14.4 Gpixels/sec | Gaussian Blur | 1.07 Gpixels/sec |
| 1.89 Gpixels/sec | Feature Matching | 0.5 Gpixels/sec |
| 1150 Gpixels/sec | Stereo Matching | 163 Gpixels/sec |
| 29783 FPS | Particle Physics | 5820.4 FPS |
| OpenCL | API | OpenCL |
| 28298 | G3D Mark Score | 10431 |
| 1094 | G2D Mark | 1094 |
| 265 FPS | DirectX 11 | 79 FPS |
| 136 FPS | DirectX 12 | 46 FPS |
| 11559 Ops/s | GPU Compute | 5066 Ops/s |
| 30250 | GB6 ML Single Precision | — |
| 52126 | GB6 ML Half Precision | — |
| 23496 | GB6 ML Quantized | — |
| 12121 | Image Classification (SP) | — |
| 33054 | Image Segmentation (HP) | — |
| 33652 | Image Super Resolution (Q) | — |
| 67213 | Face Detection (HP) | — |
| 151395 | Pose Estimation (Q) | — |
| 3209 | Text Classification (SP) | — |
| 6292 | Machine Translation (HP) | — |
| 15618 | Object Detection (SP) | — |
| 57894 | Depth Estimation (Q) | — |
| 272117 | Style Transfer (SP) | — |
| ONNX | Framework | — |
| DirectML | Backend | — |
| 170-242 GPixel/s | Pixel Fill Rate | 90 GPixel/s |
| 497-708 GTexel/s | Texture Fill Rate | 179 GTexel/s |
| 31.8-45.3 TFLOPS | FLOPS (FP32) | 5.7 TFLOPS |
| GDDR7 | Memory Type | System Shared |
| 24 GB | Memory Size | — |
| 1750 MHz | Memory Clock | — |
| 28000 Mbps | Effective Memory Speed | — |
| 256-bit | Bus | — |
| No | ECC | — |
| 896 GB/s | Memory Bandwidth | — |
| PCIe 5.0 x16 | Interface | PCIe 5.0 x16 |
| 95-175 W (configurable) | TGP | 45-120 W (configurable) |
| TSMC | Manufacturing | TSMC |
| 4 nm | Fabrication Process | 4 nm |
| 378 mm² | Die Size | 308 mm² |
| 45.6 billion | Transistor Count | — |
| 120.63 MTr/mm² | Transistor Density | — |
| 12 | DirectX | 12.2 |
| 1.4 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.1 |
| 10.1 | CUDA | No |
| Yes | Ray Tracing | Yes |
| DLSS 4 | DLSS | No |
| 2.1a | DisplayPort | 2.1a |
| Nvidia | Vendor | Amd |
| Discrete | Build | Integrated |
| January 6, 2025 | Released | January 7, 2025 |
| Laptop | Case | Laptop |
| Gaming | Purpose | Generic |
| High-end | Segment | Mid-range |
| Blackwell 2.0 | Architecture | RDNA 3.5 |
| GB203-400 | GPU Codename | Strix Halo |
| -Intel Core Ultra 9 275HXor above | Recommended CPU | -AMD Ryzen AI Max Pro 380or above |
| 26140 | Steel Nomad Lite Score | — |
| 25013 | Time Spy | — |
| 115677 | Solar Bay | — |
| 16533 | Port Royal | — |
| 51096 | Fire Strike | — |
| 47099 | Wild Life Extreme | — |
| 177139 | Night Raid | — |
| 990-1515 MHz (configurable) | Base Clock | 1295 MHz |
| 1515-2160 MHz (configurable) | Boost Clock | 2800 MHz |
| 10496 | Shading Units | 1024 |
| 328 | Texture Mapping Units (TMUs) | 64 |
| 112 | Render Output Units (ROPs) | 32 |
| 82 | Compute Units (Pipelines) | 16 |
| 328 | Tensor Cores | No |
| 82 | Ray-tracing Cores | 16 |
| 128KB per cluster | L1 Cache | 512KB per cluster |
| 64MB shared | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |