GeForce RTX 5070
Score
Nvidia RTX Pro | Gaming | |
| Workstation | |
| AI/ML | |
| Energy Efficiency | |
| Quick Comparison Final Score | |
GeForce RTX 5070
Summary
Nvidia RTX Pro| GB205-200 | Graphics Processor | GB202-300-A1 |
| 5888 | Cores | 24064 |
| 184 / 80 | TMUs / ROPs | 752 / 190 |
GeForce RTX 5070
Blender
Nvidia RTX Pro| 4935.31 | Blender GPU | 16644.8 |
GeForce RTX 5070
Games
Nvidia RTX Pro| 114 | Forza Horizon 5 | 230 |
| 156 | The Witcher 3 | 304 |
| 208 | Counter-Strike 2 | 259 |
| 133 | Far Cry 6 | 198 |
| 97 | Hogwarts Legacy | 192 |
| 145 | Call of Duty | 301 |
| 88 | Ghost of Tsushima | 172 |
| 115 | Cyberpunk 2077 | 221 |
| 194 | Tomb Raider | 303 |
| 139 | Average FPS | 242 |
GeForce RTX 5070
Average FPS by Resolution
Nvidia RTX Pro| 165 | 1080p High | 282 |
| 139 | 1080p Ultra | 242 |
| 106 | 1440p Ultra | 197 |
| 62 | 4K Ultra | 118 |
| Low | Margin of Error | High |
GeForce RTX 5070
GeekBench 6 OpenCL
Nvidia RTX Pro| 166122 | GB6 Compute Score | 419470 |
| 248 img/sec | Background Blur | 366.1 img/sec |
| 147.7 img/sec | Face Detection | 248 img/sec |
| 6.54 Gpixels/sec | Horizon Detection | 23.3 Gpixels/sec |
| 8.24 Gpixels/sec | Edge Detection | 35.2 Gpixels/sec |
| 9.55 Gpixels/sec | Gaussian Blur | 45 Gpixels/sec |
| 1.64 Gpixels/sec | Feature Matching | 2.47 Gpixels/sec |
| 876 Gpixels/sec | Stereo Matching | 3040 Gpixels/sec |
| 24026.1 FPS | Particle Physics | 42574.6 FPS |
| OpenCL | API | OpenCL |
GeForce RTX 5070
Passmark Graphics
Nvidia RTX Pro| 22633 | G3D Mark Score | 37305 |
| 967 | G2D Mark | 1375 |
| 236 FPS | DirectX 11 | 338 FPS |
| 101 FPS | DirectX 12 | 153 FPS |
| 8478 Ops/s | GPU Compute | 26895 Ops/s |
GeForce RTX 5070
GeekBench 6 ML
Nvidia RTX Pro| 25374 | GB6 ML Single Precision | 50053 |
| 46076 | GB6 ML Half Precision | 74036 |
| 19458 | GB6 ML Quantized | 37636 |
| 10408 | Image Classification (SP) | 16903 |
| 29018 | Image Segmentation (HP) | 48518 |
| 26983 | Image Super Resolution (Q) | 54618 |
| 54398 | Face Detection (HP) | 103224 |
| 107270 | Pose Estimation (Q) | 338587 |
| 3658 | Text Classification (SP) | 3894 |
| 6237 | Machine Translation (HP) | 7291 |
| 13717 | Object Detection (SP) | 26498 |
| 49509 | Depth Estimation (Q) | 90853 |
| 199816 | Style Transfer (SP) | 635289 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
GeForce RTX 5070
Raw Performance
Nvidia RTX Pro| 116-182 GPixel/s | Pixel Fill Rate | 497 GPixel/s |
| 266-420 GTexel/s | Texture Fill Rate | 1968 GTexel/s |
| 17-26.8 TFLOPS | FLOPS (FP32) | 126 TFLOPS |
GeForce RTX 5070
Memory
Nvidia RTX Pro| GDDR7 | Memory Type | GDDR7 |
| 12 GB | Memory Size | 96 GB |
| 1750 MHz | Memory Clock | 1750 MHz |
| 28000 Mbps | Effective Memory Speed | 28000 Mbps |
| 192-bit | Bus | 512-bit |
| No | ECC | Yes |
| 672 GB/s | Memory Bandwidth | 1792 GB/s |
GeForce RTX 5070
Physical
Nvidia RTX Pro| PCIe 5.0 x16 | Interface | PCIe 5.0 x16 |
| 60-140 W (configurable) | TGP | 600 W |
| TSMC | Manufacturing | TSMC |
| 4 nm | Fabrication Process | 5 nm |
| 263 mm² | Die Size | 750 mm² |
| 31.1 billion | Transistor Count | 92.2 billion |
| 118.25 MTr/mm² | Transistor Density | 122.93 MTr/mm² |
GeForce RTX 5070
API
Nvidia RTX Pro| 12.2 | DirectX | 12.2 |
| 1.4 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 12 | CUDA | 12 |
| Yes | Ray Tracing | Yes |
| DLSS 4 | DLSS | DLSS 4 |
| 2.1b | DisplayPort | 2.1b |
GeForce RTX 5070
General
Nvidia RTX Pro| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| January 7, 2025 | Released | March 18, 2026 |
| Laptop | Case | Desktop |
| Gaming | Purpose | Professional |
| Mid-range | Segment | High-end |
| Blackwell 2.0 | Architecture | Blackwell 2.0 |
| GB205-200 | GPU Codename | GB202-300-A1 |
| -Intel Core Ultra 7 255HXor above | Recommended CPU | -AMD Ryzen 9 9955HX3Dor above |
GeForce RTX 5070
3D Mark
Nvidia RTX Pro| 18210 | Steel Nomad Lite Score | 50516 |
| 17382 | Time Spy | 42437 |
| 77747 | Solar Bay | 258808 |
| 11320 | Port Royal | 39952 |
| 41818 | Fire Strike | 72183 |
| 33104 | Wild Life Extreme | 110070 |
| 158023 | Night Raid | 216213 |
GeForce RTX 5070
Graphics Processing Unit
Nvidia RTX Pro| 847-1762 MHz (configurable) | Base Clock | 1590 MHz |
| 1447-2280 MHz (configurable) | Boost Clock | 2617 MHz |
| 5888 | Shading Units | 24064 |
| 184 | Texture Mapping Units (TMUs) | 752 |
| 80 | Render Output Units (ROPs) | 190 |
| 46 | Compute Units (Pipelines) | 188 |
| 184 | Tensor Cores | 752 |
| 46 | 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 |
