Quick/Comparisons
The Nvidia RTX Pro 6000 Blackwell's exceptional gaming performance is unmatched by the Radeon RX 590, with a 84-point gap in this category.
The Nvidia RTX Pro 6000 Blackwell's perfect score in workstation performance makes it the clear winner for content creation tasks, thanks to its strong performance and 75-point gap over the Radeon RX 590.
The Nvidia RTX Pro 6000 Blackwell's capable AI/ML compute capabilities, including its 76-point lead over the Radeon RX 590, make it the winner for AI & Machine Learning tasks.
99 | Gaming | 15 |
100 | Workstation | 25 |
93 | AI/ML | 17 |
49 | Energy Efficiency | 28 |
98 | Quick Comparison Final Score | 19 |
| GB202-300-A1 | Graphics Processor | Polaris 30 |
| 24064 | Cores | 2304 |
| 752 / 190 | TMUs / ROPs | 144 / 32 |
| 16644.8 | Blender GPU | 487.92 |
| 230 | Forza Horizon 5 | 42 |
| 304 | The Witcher 3 | 47 |
| 259 | Counter-Strike 2 | 64 |
| 198 | Far Cry 6 | 50 |
| 192 | Hogwarts Legacy | 29 |
| 301 | Call of Duty | 42 |
| 172 | Ghost of Tsushima | 25 |
| 221 | Cyberpunk 2077 | 30 |
| 303 | Tomb Raider | 55 |
| 242 | Average FPS | 43 |
| 282 | 1080p High | 57 |
| 242 | 1080p Ultra | 43 |
| 197 | 1440p Ultra | 28 |
| 118 | 4K Ultra | 15 |
| High | Margin of Error | Low |
| 419470 | GB6 Compute Score | 45783 |
| 366.1 img/sec | Background Blur | 69 img/sec |
| 248 img/sec | Face Detection | 38.2 img/sec |
| 23.3 Gpixels/sec | Horizon Detection | 1.75 Gpixels/sec |
| 35.2 Gpixels/sec | Edge Detection | 3.31 Gpixels/sec |
| 45 Gpixels/sec | Gaussian Blur | 3.31 Gpixels/sec |
| 2.47 Gpixels/sec | Feature Matching | 0.38 Gpixels/sec |
| 3040 Gpixels/sec | Stereo Matching | 180.5 Gpixels/sec |
| 42574.6 FPS | Particle Physics | 6238 FPS |
| OpenCL | API | OpenCL |
| 37305 | G3D Mark Score | 9293 |
| 1375 | G2D Mark | 790 |
| 338 FPS | DirectX 11 | 65 FPS |
| 153 FPS | DirectX 12 | 45 FPS |
| 26895 Ops/s | GPU Compute | 3764 Ops/s |
| 50053 | GB6 ML Single Precision | 9064 |
| 74036 | GB6 ML Half Precision | 9971 |
| 37636 | GB6 ML Quantized | 6293 |
| 16903 | Image Classification (SP) | 3307 |
| 48518 | Image Segmentation (HP) | 4895 |
| 54618 | Image Super Resolution (Q) | 9072 |
| 103224 | Face Detection (HP) | 12467 |
| 338587 | Pose Estimation (Q) | 43468 |
| 3894 | Text Classification (SP) | 1547 |
| 7291 | Machine Translation (HP) | 2504 |
| 26498 | Object Detection (SP) | 3746 |
| 90853 | Depth Estimation (Q) | 15075 |
| 635289 | Style Transfer (SP) | 96813 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 497 GPixel/s | Pixel Fill Rate | 49 GPixel/s |
| 1968 GTexel/s | Texture Fill Rate | 222 GTexel/s |
| 126 TFLOPS | FLOPS (FP32) | 7.1 TFLOPS |
| GDDR7 | Memory Type | GDDR5 |
| 96 GB | Memory Size | 8 GB |
| 1750 MHz | Memory Clock | 2000 MHz |
| 28000 Mbps | Effective Memory Speed | 8000 Mbps |
| 512-bit | Bus | 256-bit |
| Yes | ECC | No |
| 1792 GB/s | Memory Bandwidth | 256 GB/s |
| PCIe 5.0 x16 | Interface | PCIe 3.0 x16 |
| 600 W | TGP | 175 W |
| TSMC | Manufacturing | GlobalFoundries |
| 5 nm | Fabrication Process | 12 nm |
| 750 mm² | Die Size | 232 mm² |
| 92.2 billion | Transistor Count | 5 billion |
| 122.93 MTr/mm² | Transistor Density | 21.55 MTr/mm² |
| 12.2 | DirectX | 12 |
| 1.4 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.1 |
| 12 | CUDA | — |
| Yes | Ray Tracing | No |
| DLSS 4 | DLSS | No |
| 2.1b | DisplayPort | 1.4a |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| March 18, 2026 | Released | November 15, 2018 |
| Desktop | Case | Desktop |
| Professional | Purpose | Gaming |
| High-end | Segment | High-end |
| Blackwell 2.0 | Architecture | GCN 4.0 |
| GB202-300-A1 | GPU Codename | Polaris 30 |
| -AMD Ryzen 9 9955HX3Dor above | Recommended CPU | -Intel Core i5 8500or above |
| 50516 | Steel Nomad Lite Score | 4611 |
| 42437 | Time Spy | 4778 |
| 258808 | Solar Bay | — |
| 39952 | Port Royal | — |
| 72183 | Fire Strike | 16527 |
| 110070 | Wild Life Extreme | 9735 |
| 216213 | Night Raid | 62048 |
| 1590 MHz | Base Clock | 1469 MHz |
| 2617 MHz | Boost Clock | 1545 MHz |
| 24064 | Shading Units | 2304 |
| 752 | Texture Mapping Units (TMUs) | 144 |
| 190 | Render Output Units (ROPs) | 32 |
| 188 | Compute Units (Pipelines) | — |
| 752 | Tensor Cores | No |
| 188 | Ray-tracing Cores | No |
| 128KB per cluster | L1 Cache | 16KB per cluster |
| 128MB shared | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |
Compare with something else?
Compare Nvidia RTX Pro with another →