Quick/Comparisons
46 | Gaming | 43 |
49 | Workstation | 44 |
39 | AI/ML |
Compare with something else?
Compare GeForce RTX 3070 with another →36 |
35 | Energy Efficiency | 58 |
46 | Quick Comparison Final Score | 43 |
| GA104 | Graphics Processor | GB206-250 |
| 6144 | Cores | 3840 |
| 192 / 96 | TMUs / ROPs | 120 / 48 |
| 3746.83 | Blender GPU | 3690.35 |
| 104 | Forza Horizon 5 | 94 |
| 143 | The Witcher 3 | 137 |
| 198 | Counter-Strike 2 | 188 |
| 125 | Far Cry 6 | 116 |
| 85 | Hogwarts Legacy | 79 |
| 120 | Call of Duty | 108 |
| 81 | Ghost of Tsushima | 71 |
| 102 | Cyberpunk 2077 | 89 |
| 161 | Tomb Raider | 141 |
| 124 | Average FPS | 114 |
| 150 | 1080p High | 136 |
| 124 | 1080p Ultra | 114 |
| 93 | 1440p Ultra | 85 |
| 54 | 4K Ultra | 48 |
| Low | Margin of Error | Low |
| 136096 | GB6 Compute Score | 127848 |
| 197 img/sec | Background Blur | 209.1 img/sec |
| 128.1 img/sec | Face Detection | 134.3 img/sec |
| 6.56 Gpixels/sec | Horizon Detection | 4.67 Gpixels/sec |
| 8.81 Gpixels/sec | Edge Detection | 6.13 Gpixels/sec |
| 6.47 Gpixels/sec | Gaussian Blur | 6.39 Gpixels/sec |
| 1.26 Gpixels/sec | Feature Matching | 1.36 Gpixels/sec |
| 609.3 Gpixels/sec | Stereo Matching | 607.8 Gpixels/sec |
| 18205 FPS | Particle Physics | 18827.5 FPS |
| OpenCL | API | OpenCL |
| 23260 | G3D Mark Score | 20818 |
| 1061 | G2D Mark | 1160 |
| 192 FPS | DirectX 11 | 201 FPS |
| 91 FPS | DirectX 12 | 77 FPS |
| 11452 Ops/s | GPU Compute | 10494 Ops/s |
| 22516 | GB6 ML Single Precision | 20375 |
| 36377 | GB6 ML Half Precision | 37358 |
| 14215 | GB6 ML Quantized | 15190 |
| 8615 | Image Classification (SP) | 8362 |
| 24659 | Image Segmentation (HP) | 22831 |
| 20631 | Image Super Resolution (Q) | 22375 |
| 43834 | Face Detection (HP) | 41955 |
| 98347 | Pose Estimation (Q) | 84685 |
| 3162 | Text Classification (SP) | 3296 |
| 4911 | Machine Translation (HP) | 5083 |
| 11498 | Object Detection (SP) | 11023 |
| 34719 | Depth Estimation (Q) | 38817 |
| 196833 | Style Transfer (SP) | 155103 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 170 GPixel/s | Pixel Fill Rate | 120 GPixel/s |
| 340 GTexel/s | Texture Fill Rate | 300 GTexel/s |
| 21.7 TFLOPS | FLOPS (FP32) | 19.2 TFLOPS |
| GDDR6X | Memory Type | GDDR7 |
| 8 GB | Memory Size | 8 GB |
| 1188 MHz | Memory Clock | 1750 MHz |
| 19000 Mbps | Effective Memory Speed | 28000 Mbps |
| 256-bit | Bus | 128-bit |
| No | ECC | No |
| 608.3 GB/s | Memory Bandwidth | 448 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x8 |
| 290 W | TGP | 145 W |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 5 nm |
| 392 mm² | Die Size | 181 mm² |
| 17 billion | Transistor Count | 21.9 billion |
| 43.37 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 |
| 8.6 | CUDA | 12 |
| Yes | Ray Tracing | Yes |
| DLSS 2 | DLSS | DLSS 4 |
| 1.4a | DisplayPort | 2.1b |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| June 10, 2021 | Released | April 15, 2025 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | Mid-range |
| Ampere | Architecture | Blackwell 2.0 |
| GA104 | GPU Codename | GB206-250 |
| -Intel Core i7 12700Kor above | Recommended CPU | -Intel Core Ultra 5 245Kor above |
| 15402 | Steel Nomad Lite Score | 13971 |
| 14880 | Time Spy | 13727 |
| 69900 | Solar Bay | 64441 |
| 8884 | Port Royal | 8642 |
| 37238 | Fire Strike | 37620 |
| 31911 | Wild Life Extreme | 26155 |
| 134012 | Night Raid | 133305 |
| 1575 MHz | Base Clock | 2280 MHz |
| 1770 MHz | Boost Clock | 2500 MHz |
| 6144 | Shading Units | 3840 |
| 192 | Texture Mapping Units (TMUs) | 120 |
| 96 | Render Output Units (ROPs) | 48 |
| 48 | Compute Units (Pipelines) | 30 |
| 192 | Tensor Cores | 120 |
| 48 | Ray-tracing Cores | 30 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 4MB shared | L2 Cache | 32MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |