Quick/Comparisons
70 | Gaming | 50 |
72 | Workstation | 49 |
59 | AI/ML |
Compare with something else?
Compare GeForce RTX 4070 with another →44 |
46 | Energy Efficiency | 45 |
69 | Quick Comparison Final Score | 49 |
| AD103 | Graphics Processor | Navi 32 |
| 8448 | Cores | 3456 |
| 264 / 96 | TMUs / ROPs | 216 / 96 |
| 7089.35 | Blender GPU | 2097.78 |
| 150 | Forza Horizon 5 | 119 |
| 198 | The Witcher 3 | 157 |
| 220 | Counter-Strike 2 | 197 |
| 152 | Far Cry 6 | 127 |
| 118 | Hogwarts Legacy | 90 |
| 188 | Call of Duty | 142 |
| 111 | Ghost of Tsushima | 92 |
| 155 | Cyberpunk 2077 | 103 |
| 249 | Tomb Raider | 176 |
| 171 | Average FPS | 134 |
| 200 | 1080p High | 161 |
| 171 | 1080p Ultra | 134 |
| 134 | 1440p Ultra | 99 |
| 79 | 4K Ultra | 57 |
| Low | Margin of Error | Low |
| 237068 | GB6 Compute Score | 132843 |
| 331.7 img/sec | Background Blur | 285.8 img/sec |
| 211.6 img/sec | Face Detection | 181.6 img/sec |
| 9.63 Gpixels/sec | Horizon Detection | 4.9 Gpixels/sec |
| 13.7 Gpixels/sec | Edge Detection | 6.66 Gpixels/sec |
| 13.9 Gpixels/sec | Gaussian Blur | 5.68 Gpixels/sec |
| 2.15 Gpixels/sec | Feature Matching | 1.18 Gpixels/sec |
| 1210 Gpixels/sec | Stereo Matching | 493 Gpixels/sec |
| 33566 FPS | Particle Physics | 19410.5 FPS |
| OpenCL | API | OpenCL |
| 31798 | G3D Mark Score | 22674 |
| 1240 | G2D Mark | 1160 |
| 277 FPS | DirectX 11 | 232 FPS |
| 119 FPS | DirectX 12 | 82 FPS |
| 18264 Ops/s | GPU Compute | 12978 Ops/s |
| 32793 | GB6 ML Single Precision | 25719 |
| 47412 | GB6 ML Half Precision | 30075 |
| 24566 | GB6 ML Quantized | 19789 |
| 12970 | Image Classification (SP) | 11284 |
| 26364 | Image Segmentation (HP) | 14150 |
| 37217 | Image Super Resolution (Q) | 27972 |
| 58720 | Face Detection (HP) | 35311 |
| 177604 | Pose Estimation (Q) | 142235 |
| 3815 | Text Classification (SP) | 3836 |
| 5829 | Machine Translation (HP) | 4471 |
| 15675 | Object Detection (SP) | 12084 |
| 62447 | Depth Estimation (Q) | 49197 |
| 357918 | Style Transfer (SP) | 301429 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 251 GPixel/s | Pixel Fill Rate | 244 GPixel/s |
| 689 GTexel/s | Texture Fill Rate | 550 GTexel/s |
| 44.1 TFLOPS | FLOPS (FP32) | 35.2 TFLOPS |
| GDDR6X | Memory Type | GDDR6 |
| 16 GB | Memory Size | 12 GB |
| 1313 MHz | Memory Clock | 2250 MHz |
| 21000 Mbps | Effective Memory Speed | 18000 Mbps |
| 256-bit | Bus | 192-bit |
| No | ECC | No |
| 672.3 GB/s | Memory Bandwidth | 432 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 285 W | TGP | 245 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 5 nm |
| 379 mm² | Die Size | 346 mm² |
| 45 billion | Transistor Count | 28 billion |
| 118.73 MTr/mm² | Transistor Density | 80.92 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.2 |
| 8.9 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | No |
| 1.4a | DisplayPort | 2.1 |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| January 24, 2024 | Released | September 6, 2023 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| High-end | Segment | High-end |
| Ada Lovelace | Architecture | RDNA 3.0 |
| AD103 | GPU Codename | Navi 32 |
| -Intel Core i7 14700Kor above | Recommended CPU | -Intel Core i7 13700Kor above |
| 25741 | Steel Nomad Lite Score | 14950 |
| 24222 | Time Spy | 17090 |
| 116542 | Solar Bay | 64690 |
| 15802 | Port Royal | 9119 |
| 56340 | Fire Strike | 42701 |
| 49901 | Wild Life Extreme | 30109 |
| 176609 | Night Raid | 147080 |
| 2340 MHz | Base Clock | 1435 MHz |
| 2610 MHz | Boost Clock | 2544 MHz |
| 8448 | Shading Units | 3456 |
| 264 | Texture Mapping Units (TMUs) | 216 |
| 96 | Render Output Units (ROPs) | 96 |
| 66 | Compute Units (Pipelines) | — |
| 264 | Tensor Cores | — |
| 66 | Ray-tracing Cores | 54 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 48MB shared | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 4 IPC |