Quick/Comparisons
76 | Gaming | 51 |
74 | Workstation | 56 |
66 | AI/ML |
Compare with something else?
Compare GeForce RTX 5070 with another →33 |
49 | Energy Efficiency | 41 |
74 | Quick Comparison Final Score | 49 |
| GB203-200-A1 | Graphics Processor | GA103 |
| 8960 | Cores | 7424 |
| 280 / 96 | TMUs / ROPs | 232 / 96 |
| 7584.67 | Blender GPU | 2925.42 |
| 166 | Forza Horizon 5 | — |
| 213 | The Witcher 3 | — |
| 219 | Counter-Strike 2 | — |
| 159 | Far Cry 6 | — |
| 135 | Hogwarts Legacy | — |
| 205 | Call of Duty | — |
| 121 | Ghost of Tsushima | — |
| 174 | Cyberpunk 2077 | — |
| 260 | Tomb Raider | — |
| 184 | Average FPS | — |
| 217 | 1080p High | — |
| 184 | 1080p Ultra | — |
| 148 | 1440p Ultra | — |
| 86 | 4K Ultra | — |
| Low | Margin of Error | — |
| 240508 | GB6 Compute Score | 144566 |
| 297.9 img/sec | Background Blur | 222.1 img/sec |
| 218.6 img/sec | Face Detection | 132.2 img/sec |
| 10.1 Gpixels/sec | Horizon Detection | 6.83 Gpixels/sec |
| 13.9 Gpixels/sec | Edge Detection | 9.64 Gpixels/sec |
| 14.8 Gpixels/sec | Gaussian Blur | 7.81 Gpixels/sec |
| 2.03 Gpixels/sec | Feature Matching | 1.27 Gpixels/sec |
| 1290 Gpixels/sec | Stereo Matching | 615.9 Gpixels/sec |
| 35438.3 FPS | Particle Physics | 18134.8 FPS |
| OpenCL | API | OpenCL |
| 32425 | G3D Mark Score | 17167 |
| 1329 | G2D Mark | 713 |
| 304 FPS | DirectX 11 | 132 FPS |
| 126 FPS | DirectX 12 | 84 FPS |
| 18089 Ops/s | GPU Compute | 7423 Ops/s |
| 36504 | GB6 ML Single Precision | 15704 |
| 64459 | GB6 ML Half Precision | 26550 |
| 27312 | GB6 ML Quantized | 9885 |
| 14913 | Image Classification (SP) | 6982 |
| 45199 | Image Segmentation (HP) | 15050 |
| 39360 | Image Super Resolution (Q) | 14412 |
| 84188 | Face Detection (HP) | 27475 |
| 161400 | Pose Estimation (Q) | 66491 |
| 4010 | Text Classification (SP) | 2852 |
| 7973 | Machine Translation (HP) | 3306 |
| 18934 | Object Detection (SP) | 8121 |
| 67761 | Depth Estimation (Q) | 24442 |
| 335879 | Style Transfer (SP) | 156421 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 235 GPixel/s | Pixel Fill Rate | 101-160 GPixel/s |
| 687 GTexel/s | Texture Fill Rate | 244-386 GTexel/s |
| 43.9 TFLOPS | FLOPS (FP32) | 15.6-24.7 TFLOPS |
| GDDR7 | Memory Type | GDDR6 |
| 16 GB | Memory Size | 16 GB |
| 1750 MHz | Memory Clock | 2000 MHz |
| 28000 Mbps | Effective Memory Speed | 16000 Mbps |
| 256-bit | Bus | 256-bit |
| No | ECC | Yes |
| 896 GB/s | Memory Bandwidth | 512 GB/s |
| PCIe 5.0 x16 | Interface | PCIe 4.0 x16 |
| 300 W | TGP | 80-165 W (configurable) |
| TSMC | Manufacturing | Samsung |
| 4 nm | Fabrication Process | 8 nm |
| 378 mm² | Die Size | 496 mm² |
| 45.6 billion | Transistor Count | 22 billion |
| 120.63 MTr/mm² | Transistor Density | 44.35 MTr/mm² |
| 12.2 | DirectX | 12 |
| 1.4 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 12 | CUDA | 8.6 |
| Yes | Ray Tracing | Yes |
| DLSS 4 | DLSS | DLSS 2 |
| 2.1b | DisplayPort | 1.4 |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| January 7, 2025 | Released | March 30, 2022 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Professional |
| Mid-range | Segment | High-end |
| Blackwell 2.0 | Architecture | Ampere |
| GB203-200-A1 | GPU Codename | GA103 |
| -Intel Core Ultra 5 245Kor above | Recommended CPU | -Intel Core i9 12900HXor above |
| 29989 | Steel Nomad Lite Score | — |
| 27815 | Time Spy | 8880 |
| 136964 | Solar Bay | 32222 |
| 19613 | Port Royal | 5730 |
| 67382 | Fire Strike | 24599 |
| 57734 | Wild Life Extreme | 16879 |
| 189734 | Night Raid | 27155 |
| 2295 MHz | Base Clock | — |
| 2452 MHz | Boost Clock | 1050-1663 MHz (configurable) |
| 8960 | Shading Units | 7424 |
| 280 | Texture Mapping Units (TMUs) | 232 |
| 96 | Render Output Units (ROPs) | 96 |
| 70 | Compute Units (Pipelines) | 58 |
| 280 | Tensor Cores | 232 |
| 70 | Ray-tracing Cores | 58 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 48MB shared | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |