Quick/Comparisons
The Nvidia RTX Pro 6000 Blackwell's exceptional gaming performance is unmatched by the Radeon RX 9060 XT's mediocre scores.
The Nvidia RTX Pro 6000 Blackwell's perfect score in workstation performance makes it the clear winner in content creation tasks.
Although the Radeon RX 9060 XT has dedicated AI tensor cores, the Nvidia RTX Pro 6000 Blackwell's higher AI/ML score gives it the edge in this use case.
99 | Gaming | 45 |
100 | Workstation | 43 |
93 | AI/ML | 46 |
49 | Energy Efficiency | 55 |
98 | Quick Comparison Final Score | 44 |
| GB202-300-A1 | Graphics Processor | Navi 44 |
| 24064 | Cores | 2048 |
| 752 / 190 | TMUs / ROPs | 128 / 64 |
| 16644.8 | Blender GPU | 1601.83 |
| 230 | Forza Horizon 5 | 112 |
| 304 | The Witcher 3 | 148 |
| 259 | Counter-Strike 2 | 185 |
| 198 | Far Cry 6 | 126 |
| 192 | Hogwarts Legacy | 88 |
| 301 | Call of Duty | 133 |
| 172 | Ghost of Tsushima | 90 |
| 221 | Cyberpunk 2077 | 97 |
| 303 | Tomb Raider | 162 |
| 242 | Average FPS | 127 |
| 282 | 1080p High | 155 |
| 242 | 1080p Ultra | 127 |
| 197 | 1440p Ultra | 94 |
| 118 | 4K Ultra | 54 |
| High | Margin of Error | Low |
| 419470 | GB6 Compute Score | 107640 |
| 366.1 img/sec | Background Blur | 269.5 img/sec |
| 248 img/sec | Face Detection | 151 img/sec |
| 23.3 Gpixels/sec | Horizon Detection | 3.6 Gpixels/sec |
| 35.2 Gpixels/sec | Edge Detection | 4.67 Gpixels/sec |
| 45 Gpixels/sec | Gaussian Blur | 5.27 Gpixels/sec |
| 2.47 Gpixels/sec | Feature Matching | 1.08 Gpixels/sec |
| 3040 Gpixels/sec | Stereo Matching | 395.3 Gpixels/sec |
| 42574.6 FPS | Particle Physics | 13165.7 FPS |
| OpenCL | API | OpenCL |
| 37305 | G3D Mark Score | 20129 |
| 1375 | G2D Mark | 1178 |
| 338 FPS | DirectX 11 | 185 FPS |
| 153 FPS | DirectX 12 | 57 FPS |
| 26895 Ops/s | GPU Compute | 11202 Ops/s |
| 50053 | GB6 ML Single Precision | 25924 |
| 74036 | GB6 ML Half Precision | 40429 |
| 37636 | GB6 ML Quantized | 22275 |
| 16903 | Image Classification (SP) | 12380 |
| 48518 | Image Segmentation (HP) | 15402 |
| 54618 | Image Super Resolution (Q) | 32024 |
| 103224 | Face Detection (HP) | 50457 |
| 338587 | Pose Estimation (Q) | 113796 |
| 3894 | Text Classification (SP) | 4460 |
| 7291 | Machine Translation (HP) | 7481 |
| 26498 | Object Detection (SP) | 11256 |
| 90853 | Depth Estimation (Q) | 53978 |
| 635289 | Style Transfer (SP) | 235750 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 497 GPixel/s | Pixel Fill Rate | 200 GPixel/s |
| 1968 GTexel/s | Texture Fill Rate | 401 GTexel/s |
| 126 TFLOPS | FLOPS (FP32) | 25.6 TFLOPS |
| GDDR7 | Memory Type | GDDR6 |
| 96 GB | Memory Size | 16 GB |
| 1750 MHz | Memory Clock | 2518 MHz |
| 28000 Mbps | Effective Memory Speed | 20100 Mbps |
| 512-bit | Bus | 128-bit |
| Yes | ECC | No |
| 1792 GB/s | Memory Bandwidth | 320 GB/s |
| PCIe 5.0 x16 | Interface | PCIe 5.0 x16 |
| 600 W | TGP | 160 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 4 nm |
| 750 mm² | Die Size | 199 mm² |
| 92.2 billion | Transistor Count | 29.7 billion |
| 122.93 MTr/mm² | Transistor Density | 149.25 MTr/mm² |
| 12.2 | DirectX | 12.2 |
| 1.4 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.2 |
| 12 | CUDA | No |
| Yes | Ray Tracing | Yes |
| DLSS 4 | DLSS | No |
| 2.1b | DisplayPort | 2.1a |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| March 18, 2026 | Released | June 4, 2025 |
| Desktop | Case | Desktop |
| Professional | Purpose | Gaming |
| High-end | Segment | Mid-range |
| Blackwell 2.0 | Architecture | RDNA 4.0 |
| GB202-300-A1 | GPU Codename | Navi 44 |
| -AMD Ryzen 9 9955HX3Dor above | Recommended CPU | -Intel Core Ultra 5 245Kor above |
| 50516 | Steel Nomad Lite Score | 13915 |
| 42437 | Time Spy | 16405 |
| 258808 | Solar Bay | 65446 |
| 39952 | Port Royal | 9751 |
| 72183 | Fire Strike | 38969 |
| 110070 | Wild Life Extreme | 27779 |
| 216213 | Night Raid | 147967 |
| 1590 MHz | Base Clock | 1700 MHz |
| 2617 MHz | Boost Clock | 3130 MHz |
| 24064 | Shading Units | 2048 |
| 752 | Texture Mapping Units (TMUs) | 128 |
| 190 | Render Output Units (ROPs) | 64 |
| 188 | Compute Units (Pipelines) | 32 |
| 752 | Tensor Cores | 64 |
| 188 | Ray-tracing Cores | 32 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 128MB shared | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 4 IPC |
Compare with something else?
Compare Nvidia RTX Pro with another →