Quick/Comparisons
20 | Gaming | 99 |
27 | Workstation | 100 |
15 | AI/ML |
Compare with something else?
Compare GeForce GTX 1070 with another →93 |
28 | Energy Efficiency | 49 |
22 | Quick Comparison Final Score | 98 |
| GP104 | Graphics Processor | GB202-300-A1 |
| 1920 | Cores | 24064 |
| 120 / 64 | TMUs / ROPs | 752 / 190 |
| 557.8 | Blender GPU | 16644.8 |
| 53 | Forza Horizon 5 | 230 |
| 69 | The Witcher 3 | 304 |
| 96 | Counter-Strike 2 | 259 |
| 65 | Far Cry 6 | 198 |
| 39 | Hogwarts Legacy | 192 |
| 49 | Call of Duty | 301 |
| 35 | Ghost of Tsushima | 172 |
| 40 | Cyberpunk 2077 | 221 |
| 72 | Tomb Raider | 303 |
| 58 | Average FPS | 242 |
| 71 | 1080p High | 282 |
| 58 | 1080p Ultra | 242 |
| 41 | 1440p Ultra | 197 |
| 20 | 4K Ultra | 118 |
| Low | Margin of Error | High |
| 53429 | GB6 Compute Score | 419470 |
| 142.3 img/sec | Background Blur | 366.1 img/sec |
| 65.1 img/sec | Face Detection | 248 img/sec |
| 2.2 Gpixels/sec | Horizon Detection | 23.3 Gpixels/sec |
| 3.23 Gpixels/sec | Edge Detection | 35.2 Gpixels/sec |
| 3.74 Gpixels/sec | Gaussian Blur | 45 Gpixels/sec |
| 0.3 Gpixels/sec | Feature Matching | 2.47 Gpixels/sec |
| 195.6 Gpixels/sec | Stereo Matching | 3040 Gpixels/sec |
| 5161.6 FPS | Particle Physics | 42574.6 FPS |
| OpenCL | API | OpenCL |
| 13508 | G3D Mark Score | 37305 |
| 847 | G2D Mark | 1375 |
| 100 FPS | DirectX 11 | 338 FPS |
| 48 FPS | DirectX 12 | 153 FPS |
| 6118 Ops/s | GPU Compute | 26895 Ops/s |
| 8330 | GB6 ML Single Precision | 50053 |
| 8293 | GB6 ML Half Precision | 74036 |
| 6176 | GB6 ML Quantized | 37636 |
| 2295 | Image Classification (SP) | 16903 |
| 4955 | Image Segmentation (HP) | 48518 |
| 10587 | Image Super Resolution (Q) | 54618 |
| 8747 | Face Detection (HP) | 103224 |
| 51263 | Pose Estimation (Q) | 338587 |
| 1461 | Text Classification (SP) | 3894 |
| 2342 | Machine Translation (HP) | 7291 |
| 2960 | Object Detection (SP) | 26498 |
| 15275 | Depth Estimation (Q) | 90853 |
| 104907 | Style Transfer (SP) | 635289 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 108 GPixel/s | Pixel Fill Rate | 497 GPixel/s |
| 202 GTexel/s | Texture Fill Rate | 1968 GTexel/s |
| 6.5 TFLOPS | FLOPS (FP32) | 126 TFLOPS |
| GDDR5 | Memory Type | GDDR7 |
| 8 GB | Memory Size | 96 GB |
| 2002 MHz | Memory Clock | 1750 MHz |
| 8000 Mbps | Effective Memory Speed | 28000 Mbps |
| 256-bit | Bus | 512-bit |
| No | ECC | Yes |
| 256.3 GB/s | Memory Bandwidth | 1792 GB/s |
| PCIe 3.0 x16 | Interface | PCIe 5.0 x16 |
| 150 W | TGP | 600 W |
| TSMC | Manufacturing | TSMC |
| 16 nm | Fabrication Process | 5 nm |
| 314 mm² | Die Size | 750 mm² |
| 7 billion | Transistor Count | 92.2 billion |
| 22.29 MTr/mm² | Transistor Density | 122.93 MTr/mm² |
| 12 | DirectX | 12.2 |
| 1.3 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 6.1 | CUDA | 12 |
| No | Ray Tracing | Yes |
| No | DLSS | DLSS 4 |
| 1.4a | DisplayPort | 2.1b |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| January 10, 2016 | Released | March 18, 2026 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Professional |
| Mid-range | Segment | High-end |
| Pascal | Architecture | Blackwell 2.0 |
| GP104 | GPU Codename | GB202-300-A1 |
| -Intel Core i7 7700Kor above | Recommended CPU | -AMD Ryzen 9 9955HX3Dor above |
| 5857 | Steel Nomad Lite Score | 50516 |
| 6091 | Time Spy | 42437 |
| 16018 | Solar Bay | 258808 |
| 1282 | Port Royal | 39952 |
| 18706 | Fire Strike | 72183 |
| 12190 | Wild Life Extreme | 110070 |
| 72438 | Night Raid | 216213 |
| 1506 MHz | Base Clock | 1590 MHz |
| 1683 MHz | Boost Clock | 2617 MHz |
| 1920 | Shading Units | 24064 |
| 120 | Texture Mapping Units (TMUs) | 752 |
| 64 | Render Output Units (ROPs) | 190 |
| 15 | Compute Units (Pipelines) | 188 |
| No | Tensor Cores | 752 |
| No | Ray-tracing Cores | 188 |
| 48KB per cluster | L1 Cache | 128KB per cluster |
| 2MB shared | L2 Cache | 128MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |