Quick/Comparisons
The GeForce RTX 5070 Ti takes the lead in gaming performance, thanks to its higher score of 76, outpacing the Radeon RX 7900 XT's 72.
The GeForce RTX 5070 Ti edges out the Radeon RX 7900 XT in content creation, with a score of 74 to the Radeon's 70.
The GeForce RTX 5070 Ti and Radeon RX 7900 XT are neck and neck in AI and machine learning, with scores of 66 and 68, respectively.
76 | Gaming | 72 |
74 | Workstation | 70 |
66 | AI/ML | 68 |
49 | Energy Efficiency | 45 |
74 | Quick Comparison Final Score | 70 |
| GB203-200-A1 | Graphics Processor | Navi 31 |
| 8960 | Cores | 5376 |
| 280 / 96 | TMUs / ROPs | 336 / 192 |
| 7584.67 | Blender GPU | 3548.59 |
| 166 | Forza Horizon 5 | 157 |
| 213 | The Witcher 3 | 228 |
| 219 | Counter-Strike 2 | 223 |
| 159 | Far Cry 6 | 173 |
| 135 | Hogwarts Legacy | 126 |
| 205 | Call of Duty | 222 |
| 121 | Ghost of Tsushima | 129 |
| 174 | Cyberpunk 2077 | 158 |
| 260 | Tomb Raider | 247 |
| 184 | Average FPS | 185 |
| 217 | 1080p High | 219 |
| 184 | 1080p Ultra | 185 |
| 148 | 1440p Ultra | 144 |
| 86 | 4K Ultra | 85 |
| Low | Margin of Error | Low |
| 240508 | GB6 Compute Score | 203754 |
| 297.9 img/sec | Background Blur | 419.8 img/sec |
| 218.6 img/sec | Face Detection | 240.5 img/sec |
| 10.1 Gpixels/sec | Horizon Detection | 8.72 Gpixels/sec |
| 13.9 Gpixels/sec | Edge Detection | 12 Gpixels/sec |
| 14.8 Gpixels/sec | Gaussian Blur | 9.88 Gpixels/sec |
| 2.03 Gpixels/sec | Feature Matching | 1.71 Gpixels/sec |
| 1290 Gpixels/sec | Stereo Matching | 701.7 Gpixels/sec |
| 35438.3 FPS | Particle Physics | 26631.4 FPS |
| OpenCL | API | OpenCL |
| 32425 | G3D Mark Score | 29015 |
| 1329 | G2D Mark | 1241 |
| 304 FPS | DirectX 11 | 340 FPS |
| 126 FPS | DirectX 12 | 111 FPS |
| 18089 Ops/s | GPU Compute | 16851 Ops/s |
| 36504 | GB6 ML Single Precision | 39099 |
| 64459 | GB6 ML Half Precision | 47298 |
| 27312 | GB6 ML Quantized | 30875 |
| 14913 | Image Classification (SP) | 16951 |
| 45199 | Image Segmentation (HP) | 23234 |
| 39360 | Image Super Resolution (Q) | 41128 |
| 84188 | Face Detection (HP) | 67615 |
| 161400 | Pose Estimation (Q) | 237026 |
| 4010 | Text Classification (SP) | 4961 |
| 7973 | Machine Translation (HP) | 7043 |
| 18934 | Object Detection (SP) | 18737 |
| 67761 | Depth Estimation (Q) | 67746 |
| 335879 | Style Transfer (SP) | 457919 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 235 GPixel/s | Pixel Fill Rate | 460 GPixel/s |
| 687 GTexel/s | Texture Fill Rate | 804 GTexel/s |
| 43.9 TFLOPS | FLOPS (FP32) | 51.5 TFLOPS |
| GDDR7 | Memory Type | GDDR6 |
| 16 GB | Memory Size | 20 GB |
| 1750 MHz | Memory Clock | 2500 MHz |
| 28000 Mbps | Effective Memory Speed | 20000 Mbps |
| 256-bit | Bus | 320-bit |
| No | ECC | No |
| 896 GB/s | Memory Bandwidth | 800 GB/s |
| PCIe 5.0 x16 | Interface | PCIe 4.0 x16 |
| 300 W | TGP | 300 W |
| TSMC | Manufacturing | TSMC |
| 4 nm | Fabrication Process | 5 nm |
| 378 mm² | Die Size | 529 mm² |
| 45.6 billion | Transistor Count | 57 billion |
| 120.63 MTr/mm² | Transistor Density | 107.75 MTr/mm² |
| 12.2 | DirectX | 12 |
| 1.4 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.2 |
| 12 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 4 | DLSS | No |
| 2.1b | DisplayPort | 2.1 |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| January 7, 2025 | Released | December 13, 2023 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| Mid-range | Segment | High-end |
| Blackwell 2.0 | Architecture | RDNA 3.0 |
| GB203-200-A1 | GPU Codename | Navi 31 |
| -Intel Core Ultra 5 245Kor above | Recommended CPU | -Intel Core i7 13700Kor above |
| 29989 | Steel Nomad Lite Score | 24675 |
| 27815 | Time Spy | 27067 |
| 136964 | Solar Bay | 104912 |
| 19613 | Port Royal | 14358 |
| 67382 | Fire Strike | 67256 |
| 57734 | Wild Life Extreme | 48319 |
| 189734 | Night Raid | 176267 |
| 2295 MHz | Base Clock | 1387 MHz |
| 2452 MHz | Boost Clock | 2394 MHz |
| 8960 | Shading Units | 5376 |
| 280 | Texture Mapping Units (TMUs) | 336 |
| 96 | Render Output Units (ROPs) | 192 |
| 70 | Compute Units (Pipelines) | — |
| 280 | Tensor Cores | — |
| 70 | Ray-tracing Cores | 84 |
| 128KB per cluster | L1 Cache | 256KB per cluster |
| 48MB shared | L2 Cache | 6MB shared |
| 2 IPC | Instructions Per Cycle | 4 IPC |
Compare with something else?
Compare GeForce RTX 5070 with another →