Quick/Comparisons
61 | Gaming | 63 |
62 | Workstation | 64 |
54 | AI/ML |
Compare with something else?
Compare GeForce RTX 4070 with another →49 |
46 | Energy Efficiency | 50 |
60 | Quick Comparison Final Score | 60 |
| AD104 | Graphics Processor | GB203-400 |
| 7168 | Cores | 7680 |
| 224 / 80 | TMUs / ROPs | 240 / 96 |
| 6153.24 | Blender GPU | 6681.58 |
| 131 | Forza Horizon 5 | 133 |
| 176 | The Witcher 3 | 179 |
| 212 | Counter-Strike 2 | 213 |
| 145 | Far Cry 6 | 143 |
| 106 | Hogwarts Legacy | 113 |
| 165 | Call of Duty | 173 |
| 100 | Ghost of Tsushima | 106 |
| 131 | Cyberpunk 2077 | 139 |
| 226 | Tomb Raider | 234 |
| 155 | Average FPS | 159 |
| 183 | 1080p High | 188 |
| 155 | 1080p Ultra | 159 |
| 120 | 1440p Ultra | 124 |
| 70 | 4K Ultra | 73 |
| Low | Margin of Error | Low |
| 206598 | GB6 Compute Score | 190326 |
| 359.5 img/sec | Background Blur | 194.4 img/sec |
| 210 img/sec | Face Detection | 165.6 img/sec |
| 7.61 Gpixels/sec | Horizon Detection | 8.61 Gpixels/sec |
| 10.4 Gpixels/sec | Edge Detection | 12.2 Gpixels/sec |
| 11.3 Gpixels/sec | Gaussian Blur | 11.5 Gpixels/sec |
| 1.97 Gpixels/sec | Feature Matching | 1.7 Gpixels/sec |
| 969.3 Gpixels/sec | Stereo Matching | 1060 Gpixels/sec |
| 29061.4 FPS | Particle Physics | 27590.4 FPS |
| OpenCL | API | OpenCL |
| 29964 | G3D Mark Score | 26432 |
| 1193 | G2D Mark | 1093 |
| 271 FPS | DirectX 11 | 255 FPS |
| 110 FPS | DirectX 12 | 114 FPS |
| 16985 Ops/s | GPU Compute | 10546 Ops/s |
| 30706 | GB6 ML Single Precision | 25636 |
| 43851 | GB6 ML Half Precision | 42495 |
| 23776 | GB6 ML Quantized | 19809 |
| 12146 | Image Classification (SP) | 10192 |
| 23932 | Image Segmentation (HP) | 23343 |
| 36321 | Image Super Resolution (Q) | 31132 |
| 52836 | Face Detection (HP) | 46322 |
| 179788 | Pose Estimation (Q) | 134037 |
| 3371 | Text Classification (SP) | 3319 |
| 5017 | Machine Translation (HP) | 4696 |
| 15164 | Object Detection (SP) | 13038 |
| 62500 | Depth Estimation (Q) | 52089 |
| 357489 | Style Transfer (SP) | 248270 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 198 GPixel/s | Pixel Fill Rate | 144-231 GPixel/s |
| 554 GTexel/s | Texture Fill Rate | 360-578 GTexel/s |
| 35.5 TFLOPS | FLOPS (FP32) | 23-37 TFLOPS |
| GDDR6X | Memory Type | GDDR7 |
| 12 GB | Memory Size | 16 GB |
| 1313 MHz | Memory Clock | 1750 MHz |
| 21000 Mbps | Effective Memory Speed | 28000 Mbps |
| 192-bit | Bus | 256-bit |
| No | ECC | No |
| 504.2 GB/s | Memory Bandwidth | 896 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 220 W | TGP | 80-175 W (configurable) |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 4 nm |
| 294 mm² | Die Size | 378 mm² |
| 35 billion | Transistor Count | 45.6 billion |
| 119.05 MTr/mm² | Transistor Density | 120.63 MTr/mm² |
| 12 | DirectX | 12.2 |
| 1.3 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.9 | CUDA | 10.1 |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | DLSS 4 |
| 1.4a | DisplayPort | 2.1b |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| January 17, 2024 | Released | January 7, 2025 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | High-end |
| Ada Lovelace | Architecture | Blackwell 2.0 |
| AD104 | GPU Codename | GB203-400 |
| -Intel Core i7 14700Kor above | Recommended CPU | -Intel Core Ultra 9 275HXor above |
| 21568 | Steel Nomad Lite Score | 22624 |
| 20882 | Time Spy | 22022 |
| 98166 | Solar Bay | 99378 |
| 13122 | Port Royal | 14180 |
| 50092 | Fire Strike | 47109 |
| 41630 | Wild Life Extreme | 41227 |
| 164777 | Night Raid | 170132 |
| 1980 MHz | Base Clock | 975 MHz |
| 2475 MHz | Boost Clock | 1500-2410 MHz (configurable) |
| 7168 | Shading Units | 7680 |
| 224 | Texture Mapping Units (TMUs) | 240 |
| 80 | Render Output Units (ROPs) | 96 |
| 56 | Compute Units (Pipelines) | 60 |
| 224 | Tensor Cores | 240 |
| 56 | Ray-tracing Cores | 60 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 48MB shared | L2 Cache | 64MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |