Nvidia RTX Pro 6000 Blackwell vs GeForce RTX 5090
VS
Nvidia RTX Pro 6000 Blackwell
GB202-300-A1
Nvidia GeForce RTX 5090
GB202-300
We compared two discrete desktop graphics cards: the Nvidia RTX Pro 6000 Blackwell 96 GB with 188 pipelines and 24064 shaders against the 1 year and 2 months older GeForce RTX 5090 32 GB that utilizes 170 pipelines and 21760 shaders. Here you will find complete details about specs, efficiency, performance tests, and more.
Review
Gaming
Performance in DirectX, OpenCL, and Vulkan games
Workstation
Perf. in 3D modeling, video editing and rendering apps
AI/ML
Capabilities for machine learning and AI-related tasks
Energy Efficiency
Power consumption efficiency in different scenarios
NanoReview Final Score
Overall video card score
Value for money (Beta)
Key differences
Reasons to consider the Nvidia RTX Pro 6000 Blackwell
- 20% higher maximum theoretical performance (126 vs 104.8 TFLOPS)
- Includes 64 GB more video memory
- Supports error correction code (ECC) memory
Reasons to consider the GeForce RTX 5090
- Manufactured using a more efficient 4 nm process technology
- Shows 7% higher average frame rate in modern games at QHD resolution – 207 vs 194 FPS
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 306 225 215 172 |
336 247 233 187 |
| The Witcher 3 | 336 297 219 112 |
364 327 237 123 |
| Counter-Strike 2 | 354 255 194 94 |
363 265 202 99 |
| Far Cry 6 | 220 199 184 111 |
237 202 190 121 |
| Hogwarts Legacy | 230 188 147 87 |
249 199 166 97 |
| CoD: Modern Warfare III | 301 300 246 171 |
334 322 260 184 |
| Ghost of Tsushima | 204 174 161 102 |
221 185 175 113 |
| Cyberpunk 2077 | 251 223 157 72 |
261 233 160 77 |
| Shadow of the Tomb Raider | 325 294 226 121 |
329 302 238 128 |
| 1080p High | 281 | 299 |
| 1080p Ultra | 239 | 254 |
| 1440p Ultra | 194 | 207 |
| 4K Ultra | 116 | 125 |
| Margin of Error | High | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 42441 | 46873 |
| Solar Bay | 255911 | 236689 |
| Port Royal | 39879 | 38202 |
| Fire Strike | 75085 | 87569 |
| Wild Life Extreme | 110070 | 108751 |
| Night Raid | 214655 | 206648 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 366.1 img/sec | 333 img/sec |
| Face Detection | 248 img/sec | 298.5 img/sec |
| Horizon Detection | 23.3 Gpixels/sec | 21.7 Gpixels/sec |
| Edge Detection | 35.2 Gpixels/sec | 33.7 Gpixels/sec |
| Gaussian Blur | 45 Gpixels/sec | 34.5 Gpixels/sec |
| Feature Matching | 2.47 Gpixels/sec | 2.53 Gpixels/sec |
| Stereo Matching | 3040 Gpixels/sec | 2850 Gpixels/sec |
| Particle Physics | 42574.6 FPS | 59146.9 FPS |
| API | OpenCL | OpenCL |
Passmark Graphics
G3D Mark Score
| G2D Mark | 1388 | 1410 |
| DirectX 11 | 337 FPS | 340 FPS |
| DirectX 12 | 153 FPS | 179 FPS |
| GPU Compute | 26574 Ops/s | 24383 Ops/s |
Blender
Blender GPU
Artificial Intelligence Tests
Performance in machine learning and artificial intelligence tasksGeekBench 6 ML
GB6 ML Single Precision
GB6 ML Half Precision
GB6 ML Quantized
| Image Classification (SP) | 16903 | 18948 |
| Image Segmentation (HP) | 48518 | 55109 |
| Image Super Resolution (Q) | 54618 | 58478 |
| Face Detection (HP) | 103224 | 114685 |
| Pose Estimation (Q) | 338587 | 344729 |
| Text Classification (SP) | 3894 | 4464 |
| Machine Translation (HP) | 7291 | 8184 |
| Object Detection (SP) | 26498 | 27802 |
| Depth Estimation (Q) | 90853 | 85572 |
| Style Transfer (SP) | 635289 | 687080 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of Nvidia RTX Pro 6000 Blackwell and GeForce RTX 5090General
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | March 18, 2026 | January 7, 2025 |
| Launch price (MSRP) | $8559 | $1999 |
| Case | Desktop | Desktop |
| Purpose | Professional | Gaming |
| Segment | High-end | High-end |
| Architecture | Blackwell 2.0 | Blackwell 2.0 |
| GPU Codename | GB202-300-A1 | GB202-300 |
| Recommended CPU | - AMD Ryzen 9 9955HX3D or above | - Intel Core Ultra 9 285K or above |
Desktop GPU rating (1st and 2nd place)
Graphics Processing Unit
| Base Clock | 1590 MHz | 2017 MHz |
| Boost Clock | 2617 MHz | 2407 MHz |
| Shading Units | 24064 | 21760 |
| Texture Mapping Units (TMUs) | 752 | 680 |
| Render Output Units (ROPs) | 190 | 176 |
| Compute Units (Pipelines) | 188 | 170 |
| Tensor Cores | 752 | 680 |
| Ray-tracing Cores | 188 | 170 |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 128MB shared | 96MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 497 GPixel/s | 424 GPixel/s |
| Texture Fill Rate | 1968 GTexel/s | 1637 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 5.0 x16 | PCIe 5.0 x16 |
| TGP | 600 W | 575 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 5 nm | 4 nm |
| Die Size | 750 mm² | 750 mm² |
| Transistor Count | 92.2 billion | 92.2 billion |
| Transistor Density | 122.93 MTr/mm² | 122.93 MTr/mm² |
Memory
| Memory Type | GDDR7 | GDDR7 |
| Memory Size | 96 GB | 32 GB |
| Memory Clock | 1750 MHz | 1750 MHz |
| Effective Memory Speed | 28000 Mbps | 28000 Mbps |
| Bus | 512-bit | 512-bit |
| ECC | Yes | No |
Memory Bandwidth
API
| DirectX | 12.2 | 12.2 |
| Vulkan | 1.4 | 1.4 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| CUDA | 12 | 12.0 |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 4 | DLSS 4 |
| DisplayPort | 2.1b | 2.1b |
Cast your vote
Total votes: 183