Quick/Comparisons
The Nvidia RTX Pro 6000 Blackwell's superior gaming performance, with a score of 99, makes it the clear winner in this use case. Its 99 score is significantly higher than the GeForce RTX 3090's 61 score.
The Nvidia RTX Pro 6000 Blackwell's perfect score of 100 in the Content Creation use case makes it the clear winner, thanks to its strong workstation performance. The GeForce RTX 3090's 69 score is respectable but falls short of the Nvidia RTX Pro 6000 Blackwell's mark.
The Nvidia RTX Pro 6000 Blackwell's strong AI/ML compute capabilities, with a score of 93, make it the clear winner in this use case. Its 93 score is significantly higher than the GeForce RTX 3090's 53 score.
61 | Gaming | 99 |
69 | Workstation | 100 |
53 | AI/ML |
Compare with something else?
Compare GeForce RTX 3090 with another →93 |
35 | Energy Efficiency | 49 |
62 | Quick Comparison Final Score | 98 |
| GA102 | Graphics Processor | GB202-300-A1 |
| 10496 | Cores | 24064 |
| 328 / 112 | TMUs / ROPs | 752 / 190 |
| 5800.59 | Blender GPU | 16644.8 |
| 125 | Forza Horizon 5 | 230 |
| 175 | The Witcher 3 | 304 |
| 212 | Counter-Strike 2 | 259 |
| 142 | Far Cry 6 | 198 |
| 106 | Hogwarts Legacy | 192 |
| 159 | Call of Duty | 301 |
| 97 | Ghost of Tsushima | 172 |
| 132 | Cyberpunk 2077 | 221 |
| 226 | Tomb Raider | 303 |
| 153 | Average FPS | 242 |
| 179 | 1080p High | 282 |
| 153 | 1080p Ultra | 242 |
| 119 | 1440p Ultra | 197 |
| 70 | 4K Ultra | 118 |
| Low | Margin of Error | High |
| 211669 | GB6 Compute Score | 419470 |
| 320.8 img/sec | Background Blur | 366.1 img/sec |
| 207.2 img/sec | Face Detection | 248 img/sec |
| 10.1 Gpixels/sec | Horizon Detection | 23.3 Gpixels/sec |
| 14.4 Gpixels/sec | Edge Detection | 35.2 Gpixels/sec |
| 11.5 Gpixels/sec | Gaussian Blur | 45 Gpixels/sec |
| 1.48 Gpixels/sec | Feature Matching | 2.47 Gpixels/sec |
| 984.2 Gpixels/sec | Stereo Matching | 3040 Gpixels/sec |
| 27764.9 FPS | Particle Physics | 42574.6 FPS |
| OpenCL | API | OpenCL |
| 26551 | G3D Mark Score | 37305 |
| 1069 | G2D Mark | 1375 |
| 218 FPS | DirectX 11 | 338 FPS |
| 110 FPS | DirectX 12 | 153 FPS |
| 15117 Ops/s | GPU Compute | 26895 Ops/s |
| 27237 | GB6 ML Single Precision | 50053 |
| 41500 | GB6 ML Half Precision | 74036 |
| 15927 | GB6 ML Quantized | 37636 |
| 10665 | Image Classification (SP) | 16903 |
| 28963 | Image Segmentation (HP) | 48518 |
| 22697 | Image Super Resolution (Q) | 54618 |
| 55232 | Face Detection (HP) | 103224 |
| 128155 | Pose Estimation (Q) | 338587 |
| 3152 | Text Classification (SP) | 3894 |
| 5088 | Machine Translation (HP) | 7291 |
| 13891 | Object Detection (SP) | 26498 |
| 37984 | Depth Estimation (Q) | 90853 |
| 272000 | Style Transfer (SP) | 635289 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 190 GPixel/s | Pixel Fill Rate | 497 GPixel/s |
| 556 GTexel/s | Texture Fill Rate | 1968 GTexel/s |
| 35.6 TFLOPS | FLOPS (FP32) | 126 TFLOPS |
| GDDR6X | Memory Type | GDDR7 |
| 24 GB | Memory Size | 96 GB |
| 1219 MHz | Memory Clock | 1750 MHz |
| 19500 Mbps | Effective Memory Speed | 28000 Mbps |
| 384-bit | Bus | 512-bit |
| No | ECC | Yes |
| 936.2 GB/s | Memory Bandwidth | 1792 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 350 W | TGP | 600 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 5 nm |
| 628 mm² | Die Size | 750 mm² |
| 28 billion | Transistor Count | 92.2 billion |
| 44.59 MTr/mm² | Transistor Density | 122.93 MTr/mm² |
| 12 | DirectX | 12.2 |
| 1.3 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.6 | CUDA | 12 |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | DLSS 4 |
| 1.4a | DisplayPort | 2.1b |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| September 24, 2020 | Released | March 18, 2026 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Professional |
| High-end | Segment | High-end |
| Ampere | Architecture | Blackwell 2.0 |
| GA102 | GPU Codename | GB202-300-A1 |
| -Intel Core i9 12900Kor above | Recommended CPU | -AMD Ryzen 9 9955HX3Dor above |
| 21851 | Steel Nomad Lite Score | 50516 |
| 19896 | Time Spy | 42437 |
| 97375 | Solar Bay | 258808 |
| 13631 | Port Royal | 39952 |
| 47515 | Fire Strike | 72183 |
| 43673 | Wild Life Extreme | 110070 |
| 139755 | Night Raid | 216213 |
| 1395 MHz | Base Clock | 1590 MHz |
| 1695 MHz | Boost Clock | 2617 MHz |
| 10496 | Shading Units | 24064 |
| 328 | Texture Mapping Units (TMUs) | 752 |
| 112 | Render Output Units (ROPs) | 190 |
| 82 | Compute Units (Pipelines) | 188 |
| 328 | Tensor Cores | 752 |
| 82 | Ray-tracing Cores | 188 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 6MB shared | L2 Cache | 128MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |