Quick/Comparisons
43 | Gaming | 34 |
44 | Workstation | 36 |
35 | AI/ML |
Compare with something else?
Compare GeForce RTX 3070 with another →31 |
35 | Energy Efficiency | 62 |
42 | Quick Comparison Final Score | 35 |
| GA104 | Graphics Processor | GB207-300 |
| 5888 | Cores | 2560 |
| 184 / 96 | TMUs / ROPs | 80 / 32 |
| 3368.42 | Blender GPU | 2871.08 |
| 99 | Forza Horizon 5 | 75 |
| 143 | The Witcher 3 | 112 |
| 194 | Counter-Strike 2 | 150 |
| 121 | Far Cry 6 | 91 |
| 80 | Hogwarts Legacy | 61 |
| 113 | Call of Duty | 83 |
| 74 | Ghost of Tsushima | 56 |
| 94 | Cyberpunk 2077 | 72 |
| 153 | Tomb Raider | 114 |
| 119 | Average FPS | 90 |
| 141 | 1080p High | 112 |
| 119 | 1080p Ultra | 90 |
| 88 | 1440p Ultra | 67 |
| 50 | 4K Ultra | 37 |
| Low | Margin of Error | Low |
| 126195 | GB6 Compute Score | 92181 |
| 161.6 img/sec | Background Blur | 129.6 img/sec |
| 117.2 img/sec | Face Detection | 94.6 img/sec |
| 5.89 Gpixels/sec | Horizon Detection | 3.45 Gpixels/sec |
| 8.05 Gpixels/sec | Edge Detection | 4.41 Gpixels/sec |
| 6.65 Gpixels/sec | Gaussian Blur | 4.43 Gpixels/sec |
| 1.25 Gpixels/sec | Feature Matching | 1.16 Gpixels/sec |
| 550.6 Gpixels/sec | Stereo Matching | 434.1 Gpixels/sec |
| 17613.1 FPS | Particle Physics | 13990.8 FPS |
| OpenCL | API | OpenCL |
| 22128 | G3D Mark Score | 17292 |
| 1005 | G2D Mark | 1084 |
| 182 FPS | DirectX 11 | 152 FPS |
| 84 FPS | DirectX 12 | 68 FPS |
| 11062 Ops/s | GPU Compute | 8547 Ops/s |
| 19789 | GB6 ML Single Precision | 16926 |
| 33013 | GB6 ML Half Precision | 34904 |
| 12443 | GB6 ML Quantized | 13248 |
| 7624 | Image Classification (SP) | 7103 |
| 23420 | Image Segmentation (HP) | 22390 |
| 17898 | Image Super Resolution (Q) | 17642 |
| 40535 | Face Detection (HP) | 40054 |
| 83624 | Pose Estimation (Q) | 66253 |
| 2724 | Text Classification (SP) | 3436 |
| 4481 | Machine Translation (HP) | 5705 |
| 10304 | Object Detection (SP) | 8910 |
| 30418 | Depth Estimation (Q) | 31732 |
| 180089 | Style Transfer (SP) | 107993 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 166 GPixel/s | Pixel Fill Rate | 82 GPixel/s |
| 317 GTexel/s | Texture Fill Rate | 206 GTexel/s |
| 20.3 TFLOPS | FLOPS (FP32) | 13.2 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 8 GB | Memory Size | 8 GB |
| 1750 MHz | Memory Clock | 1750 MHz |
| 14000 Mbps | Effective Memory Speed | 20000 Mbps |
| 256-bit | Bus | 128-bit |
| No | ECC | No |
| 448 GB/s | Memory Bandwidth | 320 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x8 |
| 220 W | TGP | 130 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 5 nm |
| 392 mm² | Die Size | 149 mm² |
| 17 billion | Transistor Count | 16.9 billion |
| 43.37 MTr/mm² | Transistor Density | 113.42 MTr/mm² |
| 12 | DirectX | 12.2 |
| 1.3 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.6 | CUDA | 12 |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | DLSS 4 |
| 1.4a | DisplayPort | 2.1b |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| October 22, 2020 | Released | June 24, 2025 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Ampere | Architecture | Blackwell 2.0 |
| GA104 | GPU Codename | GB207-300 |
| -Intel Core i7 12700Kor above | Recommended CPU | -Intel Core Ultra 5 245Kor above |
| 14153 | Steel Nomad Lite Score | 10067 |
| 13643 | Time Spy | 10206 |
| 64426 | Solar Bay | 47184 |
| 8311 | Port Royal | 6164 |
| 34054 | Fire Strike | 28682 |
| 28486 | Wild Life Extreme | 18890 |
| 122039 | Night Raid | 111889 |
| 1500 MHz | Base Clock | 2317 MHz |
| 1725 MHz | Boost Clock | 2572 MHz |
| 5888 | Shading Units | 2560 |
| 184 | Texture Mapping Units (TMUs) | 80 |
| 96 | Render Output Units (ROPs) | 32 |
| 46 | Compute Units (Pipelines) | 20 |
| 184 | Tensor Cores | 80 |
| 46 | Ray-tracing Cores | 20 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 4MB shared | L2 Cache | 32MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |