Quick/Comparisons
66 | Gaming | 50 |
66 | Workstation | 47 |
57 | AI/ML |
Compare with something else?
Compare GeForce RTX 4070 with another →42 |
44 | Energy Efficiency | 48 |
64 | Quick Comparison Final Score | 48 |
| AD104 | Graphics Processor | GB206-300 |
| 7680 | Cores | 4608 |
| 240 / 80 | TMUs / ROPs | 144 / 48 |
| 6639.7 | Blender GPU | 4313.42 |
| 141 | Forza Horizon 5 | 104 |
| 182 | The Witcher 3 | 147 |
| 216 | Counter-Strike 2 | 205 |
| 148 | Far Cry 6 | 125 |
| 114 | Hogwarts Legacy | 90 |
| 171 | Call of Duty | 131 |
| 106 | Ghost of Tsushima | 81 |
| 142 | Cyberpunk 2077 | 106 |
| 245 | Tomb Raider | 171 |
| 163 | Average FPS | 129 |
| 194 | 1080p High | 155 |
| 163 | 1080p Ultra | 129 |
| 128 | 1440p Ultra | 99 |
| 75 | 4K Ultra | 57 |
| Low | Margin of Error | Low |
| 219163 | GB6 Compute Score | 141028 |
| 371.2 img/sec | Background Blur | 197.7 img/sec |
| 235 img/sec | Face Detection | 131.5 img/sec |
| 7.51 Gpixels/sec | Horizon Detection | 5.28 Gpixels/sec |
| 10.6 Gpixels/sec | Edge Detection | 6.79 Gpixels/sec |
| 12.4 Gpixels/sec | Gaussian Blur | 7.74 Gpixels/sec |
| 2.03 Gpixels/sec | Feature Matching | 1.53 Gpixels/sec |
| 1120 Gpixels/sec | Stereo Matching | 726.7 Gpixels/sec |
| 30420.2 FPS | Particle Physics | 21833.6 FPS |
| OpenCL | API | OpenCL |
| 31568 | G3D Mark Score | 22659 |
| 1209 | G2D Mark | 1212 |
| 288 FPS | DirectX 11 | 219 FPS |
| 116 FPS | DirectX 12 | 89 FPS |
| 18221 Ops/s | GPU Compute | 11327 Ops/s |
| 32405 | GB6 ML Single Precision | 23734 |
| 48989 | GB6 ML Half Precision | 43218 |
| 24894 | GB6 ML Quantized | 18191 |
| 12496 | Image Classification (SP) | 9511 |
| 33378 | Image Segmentation (HP) | 29012 |
| 37180 | Image Super Resolution (Q) | 25621 |
| 62919 | Face Detection (HP) | 52204 |
| 171125 | Pose Estimation (Q) | 122234 |
| 3630 | Text Classification (SP) | 3209 |
| 6420 | Machine Translation (HP) | 6074 |
| 15804 | Object Detection (SP) | 12095 |
| 58647 | Depth Estimation (Q) | 45666 |
| 331379 | Style Transfer (SP) | 182591 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 209 GPixel/s | Pixel Fill Rate | 123 GPixel/s |
| 626 GTexel/s | Texture Fill Rate | 370 GTexel/s |
| 40.1 TFLOPS | FLOPS (FP32) | 23.7 TFLOPS |
| GDDR6X | Memory Type | GDDR7 |
| 12 GB | Memory Size | 8 GB |
| 1313 MHz | Memory Clock | 1750 MHz |
| 21000 Mbps | Effective Memory Speed | 28000 Mbps |
| 192-bit | Bus | 128-bit |
| No | ECC | No |
| 504 GB/s | Memory Bandwidth | 448 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x8 |
| 285 W | TGP | 180 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 5 nm |
| 294 mm² | Die Size | 181 mm² |
| 35 billion | Transistor Count | 21.9 billion |
| 119.05 MTr/mm² | Transistor Density | 120.99 MTr/mm² |
| 12 | DirectX | 12.2 |
| 1.3 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.9 | CUDA | 12 |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | DLSS 4 |
| 1.4a | DisplayPort | 2.1b |
| Mid-range | Segment | Mid-range |
| Ada Lovelace | Architecture | Blackwell 2.0 |
| AD104 | GPU Codename | GB206-300 |
| -Intel Core i7 14700Kor above | Recommended CPU | -Intel Core Ultra 5 245Kor above |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| January 3, 2023 | Released | April 15, 2025 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| 23618 | Steel Nomad Lite Score | 16172 |
| 22751 | Time Spy | 15835 |
| 108422 | Solar Bay | 75721 |
| 13994 | Port Royal | 10077 |
| 53484 | Fire Strike | 41970 |
| 45397 | Wild Life Extreme | 30331 |
| 171258 | Night Raid | 140626 |
| 2310 MHz | Base Clock | 2407 MHz |
| 2610 MHz | Boost Clock | 2572 MHz |
| 7680 | Shading Units | 4608 |
| 240 | Texture Mapping Units (TMUs) | 144 |
| 80 | Render Output Units (ROPs) | 48 |
| 60 | Compute Units (Pipelines) | 36 |
| 240 | Tensor Cores | 144 |
| 60 | Ray-tracing Cores | 36 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 48MB shared | L2 Cache | 32MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |