Quick/Comparisons
The Nvidia RTX Pro 6000 Blackwell's exceptional gaming performance makes it the clear winner, with a 56-point gap in this category.
The Nvidia RTX Pro 6000 Blackwell's perfect score in workstation performance secures its win, thanks to its strong performance in content creation tasks.
Although the Nvidia RTX 3500 Laptop (Ada) trails behind, the Nvidia RTX Pro 6000 Blackwell's 54-point lead in AI/ML compute makes it the winner in this category.
43 | Gaming | 99 |
53 | Workstation | 100 |
39 | AI/ML | 93 |
56 | Energy Efficiency | 49 |
47 | Quick Comparison Final Score | 98 |
| AD104 | Graphics Processor | GB202-300-A1 |
| 5120 | Cores | 24064 |
| 160 / 64 | TMUs / ROPs | 752 / 190 |
| 4377.08 | Blender GPU | 16644.8 |
| 93 | Forza Horizon 5 | 230 |
| 134 | The Witcher 3 | 304 |
| 191 | Counter-Strike 2 | 259 |
| 113 | Far Cry 6 | 198 |
| 80 | Hogwarts Legacy | 192 |
| 111 | Call of Duty | 301 |
| 72 | Ghost of Tsushima | 172 |
| 89 | Cyberpunk 2077 | 221 |
| 145 | Tomb Raider | 303 |
| 114 | Average FPS | 242 |
| 138 | 1080p High | 282 |
| 114 | 1080p Ultra | 242 |
| 84 | 1440p Ultra | 197 |
| 49 | 4K Ultra | 118 |
| High | Margin of Error | High |
| 131037 | GB6 Compute Score | 419470 |
| 223 img/sec | Background Blur | 366.1 img/sec |
| 145.8 img/sec | Face Detection | 248 img/sec |
| 5.87 Gpixels/sec | Horizon Detection | 23.3 Gpixels/sec |
| 7.77 Gpixels/sec | Edge Detection | 35.2 Gpixels/sec |
| 7.2 Gpixels/sec | Gaussian Blur | 45 Gpixels/sec |
| 1.39 Gpixels/sec | Feature Matching | 2.47 Gpixels/sec |
| 458.7 Gpixels/sec | Stereo Matching | 3040 Gpixels/sec |
| 14313 FPS | Particle Physics | 42574.6 FPS |
| OpenCL | API | OpenCL |
| 19719 | G3D Mark Score | 37305 |
| 712 | G2D Mark | 1375 |
| 187 FPS | DirectX 11 | 338 FPS |
| 83 FPS | DirectX 12 | 153 FPS |
| 7623 Ops/s | GPU Compute | 26895 Ops/s |
| 20783 | GB6 ML Single Precision | 50053 |
| 33871 | GB6 ML Half Precision | 74036 |
| 16429 | GB6 ML Quantized | 37636 |
| 8316 | Image Classification (SP) | 16903 |
| 17301 | Image Segmentation (HP) | 48518 |
| 27105 | Image Super Resolution (Q) | 54618 |
| 38189 | Face Detection (HP) | 103224 |
| 105500 | Pose Estimation (Q) | 338587 |
| 3319 | Text Classification (SP) | 3894 |
| 3845 | Machine Translation (HP) | 7291 |
| 10208 | Object Detection (SP) | 26498 |
| 38369 | Depth Estimation (Q) | 90853 |
| 196359 | Style Transfer (SP) | 635289 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 99-144 GPixel/s | Pixel Fill Rate | 497 GPixel/s |
| 247-360 GTexel/s | Texture Fill Rate | 1968 GTexel/s |
| 15.8-23 TFLOPS | FLOPS (FP32) | 126 TFLOPS |
| GDDR6 | Memory Type | GDDR7 |
| 12 GB | Memory Size | 96 GB |
| 2250 MHz | Memory Clock | 1750 MHz |
| 18000 Mbps | Effective Memory Speed | 28000 Mbps |
| 192-bit | Bus | 512-bit |
| Yes | ECC | Yes |
| 432 GB/s | Memory Bandwidth | 1792 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 60-140 W (configurable) | TGP | 600 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 5 nm |
| 294 mm² | Die Size | 750 mm² |
| 35.8 billion | Transistor Count | 92.2 billion |
| 121.77 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 |
| 8.9 | CUDA | 12 |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | DLSS 4 |
| 1.4a | DisplayPort | 2.1b |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| March 21, 2023 | Released | March 18, 2026 |
| Laptop | Case | Desktop |
| Professional | Purpose | Professional |
| Mid-range | Segment | High-end |
| Ada Lovelace | Architecture | Blackwell 2.0 |
| AD104 | GPU Codename | GB202-300-A1 |
| -Intel Core Ultra 7 165Hor above | Recommended CPU | -AMD Ryzen 9 9955HX3Dor above |
| 13536 | Steel Nomad Lite Score | 50516 |
| 12966 | Time Spy | 42437 |
| 52474 | Solar Bay | 258808 |
| 7976 | Port Royal | 39952 |
| 32977 | Fire Strike | 72183 |
| 26514 | Wild Life Extreme | 110070 |
| 110011 | Night Raid | 216213 |
| — | Base Clock | 1590 MHz |
| 1545-2250 MHz (configurable) | Boost Clock | 2617 MHz |
| 5120 | Shading Units | 24064 |
| 160 | Texture Mapping Units (TMUs) | 752 |
| 64 | Render Output Units (ROPs) | 190 |
| 40 | Compute Units (Pipelines) | 188 |
| 160 | Tensor Cores | 752 |
| 40 | Ray-tracing Cores | 188 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 48MB shared | L2 Cache | 128MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |
Compare with something else?
Compare Nvidia RTX 3500 with another →