Quick/Comparisons
61 | Gaming | 34 |
69 | Workstation | 32 |
53 | AI/ML |
Compare with something else?
Compare GeForce RTX 3090 with another →21 |
35 | Energy Efficiency | 78 |
62 | Quick Comparison Final Score | 39 |
| GA102 | Graphics Processor | GB207-300-A1 |
| 10496 | Cores | 1792 |
| 328 / 112 | TMUs / ROPs | 56 / 24 |
| 5800.59 | Blender GPU | 2128.65 |
| 125 | Forza Horizon 5 | — |
| 175 | The Witcher 3 | — |
| 212 | Counter-Strike 2 | — |
| 142 | Far Cry 6 | — |
| 106 | Hogwarts Legacy | — |
| 159 | Call of Duty | — |
| 97 | Ghost of Tsushima | — |
| 132 | Cyberpunk 2077 | — |
| 226 | Tomb Raider | — |
| 153 | Average FPS | — |
| 179 | 1080p High | — |
| 153 | 1080p Ultra | — |
| 119 | 1440p Ultra | — |
| 70 | 4K Ultra | — |
| Low | Margin of Error | — |
| 211669 | GB6 Compute Score | 66956 |
| 320.8 img/sec | Background Blur | 112.8 img/sec |
| 207.2 img/sec | Face Detection | 66.5 img/sec |
| 10.1 Gpixels/sec | Horizon Detection | 2.36 Gpixels/sec |
| 14.4 Gpixels/sec | Edge Detection | 3 Gpixels/sec |
| 11.5 Gpixels/sec | Gaussian Blur | 3.67 Gpixels/sec |
| 1.48 Gpixels/sec | Feature Matching | 0.85 Gpixels/sec |
| 984.2 Gpixels/sec | Stereo Matching | 287.1 Gpixels/sec |
| 27764.9 FPS | Particle Physics | 9494.1 FPS |
| OpenCL | API | OpenCL |
| 26551 | G3D Mark Score | 13205 |
| 1069 | G2D Mark | 702 |
| 218 FPS | DirectX 11 | 99 FPS |
| 110 FPS | DirectX 12 | 69 FPS |
| 15117 Ops/s | GPU Compute | 5058 Ops/s |
| 27237 | GB6 ML Single Precision | 11766 |
| 41500 | GB6 ML Half Precision | 23687 |
| 15927 | GB6 ML Quantized | 9527 |
| 10665 | Image Classification (SP) | 4878 |
| 28963 | Image Segmentation (HP) | 15282 |
| 22697 | Image Super Resolution (Q) | 12309 |
| 55232 | Face Detection (HP) | 27767 |
| 128155 | Pose Estimation (Q) | 44436 |
| 3152 | Text Classification (SP) | 2471 |
| 5088 | Machine Translation (HP) | 3958 |
| 13891 | Object Detection (SP) | 6278 |
| 37984 | Depth Estimation (Q) | 23945 |
| 272000 | Style Transfer (SP) | 70513 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 190 GPixel/s | Pixel Fill Rate | 43-56 GPixel/s |
| 556 GTexel/s | Texture Fill Rate | 100-131 GTexel/s |
| 35.6 TFLOPS | FLOPS (FP32) | 6.4-8.4 TFLOPS |
| GDDR6X | Memory Type | GDDR7 |
| 24 GB | Memory Size | 6 GB |
| 1219 MHz | Memory Clock | 1500 MHz |
| 19500 Mbps | Effective Memory Speed | 24000 Mbps |
| 384-bit | Bus | 96-bit |
| No | ECC | No |
| 936.2 GB/s | Memory Bandwidth | 288 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 350 W | TGP | 35-75 W (configurable) |
| Samsung | Manufacturing | TSMC |
| 8 nm | Fabrication Process | 5 nm |
| 628 mm² | Die Size | 149 mm² |
| 28 billion | Transistor Count | 16.9 billion |
| 44.59 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 |
| September 24, 2020 | Released | March 18, 2025 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Professional |
| High-end | Segment | Entry-level |
| Ampere | Architecture | Blackwell 2.0 |
| GA102 | GPU Codename | GB207-300-A1 |
| -Intel Core i9 12900Kor above | Recommended CPU | -Intel Core Ultra 7 255Hor above |
| 21851 | Steel Nomad Lite Score | — |
| 19896 | Time Spy | — |
| 97375 | Solar Bay | — |
| 13631 | Port Royal | — |
| 47515 | Fire Strike | — |
| 43673 | Wild Life Extreme | — |
| 139755 | Night Raid | — |
| 1395 MHz | Base Clock | 1357-1792 MHz (configurable) |
| 1695 MHz | Boost Clock | 1792-2332 MHz (configurable) |
| 10496 | Shading Units | 1792 |
| 328 | Texture Mapping Units (TMUs) | 56 |
| 112 | Render Output Units (ROPs) | 24 |
| 82 | Compute Units (Pipelines) | 14 |
| 328 | Tensor Cores | 56 |
| 82 | Ray-tracing Cores | 14 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 6MB shared | L2 Cache | 24MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |