Quick/Comparisons
The GeForce RTX 4090 takes the lead in gaming performance, thanks to its 92 score, which is significantly higher than the GeForce RTX 5070 Laptop's 45 score.
The GeForce RTX 4090 excels in workstation performance, with a score of 91, outpacing the GeForce RTX 5070 Laptop's 45 score.
The GeForce RTX 4090 has a notable edge in AI/ML compute, with a score of 76, compared to the GeForce RTX 5070 Laptop's 39 score.
92 | Gaming | 45 |
91 | Workstation | 45 |
76 | AI/ML | 39 |
43 | Energy Efficiency | 65 |
89 | Quick Comparison Final Score | 48 |
| AD102 | Graphics Processor | GB206-300 |
| 16384 | Cores | 4608 |
| 512 / 176 | TMUs / ROPs | 144 / 48 |
| 11688.48 | Blender GPU | 4007.35 |
| 203 | Forza Horizon 5 | 96 |
| 270 | The Witcher 3 | 137 |
| 245 | Counter-Strike 2 | 195 |
| 183 | Far Cry 6 | 116 |
| 165 | Hogwarts Legacy | 79 |
| 269 | Call of Duty | 110 |
| 150 | Ghost of Tsushima | 73 |
| 204 | Cyberpunk 2077 | 92 |
| 283 | Tomb Raider | 149 |
| 219 | Average FPS | 116 |
| 256 | 1080p High | 140 |
| 219 | 1080p Ultra | 116 |
| 176 | 1440p Ultra | 88 |
| 107 | 4K Ultra | 51 |
| Low | Margin of Error | Low |
| 2.38 Gpixels/sec | Feature Matching | 1.46 Gpixels/sec |
| 1950 Gpixels/sec | Stereo Matching | 661.9 Gpixels/sec |
| 48090.4 FPS | Particle Physics | 19395.6 FPS |
| OpenCL | API | OpenCL |
| 316301 | GB6 Compute Score | 134209 |
| 318.9 img/sec | Background Blur | 213.4 img/sec |
| 222.8 img/sec | Face Detection | 130.4 img/sec |
| 14.2 Gpixels/sec | Horizon Detection | 4.89 Gpixels/sec |
| 21 Gpixels/sec | Edge Detection | 6.23 Gpixels/sec |
| 23.9 Gpixels/sec | Gaussian Blur | 7.41 Gpixels/sec |
| 38071 | G3D Mark Score | 19173 |
| 1303 | G2D Mark | 884 |
| 324 FPS | DirectX 11 | 186 FPS |
| 151 FPS | DirectX 12 | 91 FPS |
| 26193 Ops/s | GPU Compute | 7114 Ops/s |
| 42962 | GB6 ML Single Precision | 22022 |
| 59636 | GB6 ML Half Precision | 40929 |
| 31670 | GB6 ML Quantized | 17069 |
| 15600 | Image Classification (SP) | 8984 |
| 36657 | Image Segmentation (HP) | 26270 |
| 48114 | Image Super Resolution (Q) | 24606 |
| 78111 | Face Detection (HP) | 48283 |
| 289639 | Pose Estimation (Q) | 98180 |
| 4074 | Text Classification (SP) | 3260 |
| 6349 | Machine Translation (HP) | 5770 |
| 21186 | Object Detection (SP) | 11172 |
| 75599 | Depth Estimation (Q) | 43305 |
| 601428 | Style Transfer (SP) | 165555 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 444 GPixel/s | Pixel Fill Rate | 68-121 GPixel/s |
| 1290 GTexel/s | Texture Fill Rate | 203-363 GTexel/s |
| 82.6 TFLOPS | FLOPS (FP32) | 13-23.2 TFLOPS |
| GDDR6X | Memory Type | GDDR7 |
| 24 GB | Memory Size | 8 GB |
| 2625 MHz | Memory Clock | 1750 MHz |
| 21000 Mbps | Effective Memory Speed | 24000 Mbps |
| 384-bit | Bus | 128-bit |
| No | ECC | No |
| 1010 GB/s | Memory Bandwidth | 384 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 450 W | TGP | 50-125 W (configurable) |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 4 nm |
| 609 mm² | Die Size | 181 mm² |
| 76.3 billion | Transistor Count | 21.9 billion |
| 125.29 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 |
| 8.9 | CUDA | 12 |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | DLSS 4 |
| 1.4a | DisplayPort | 2.1b |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| September 20, 2022 | Released | January 7, 2025 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Gaming |
| High-end | Segment | Mid-range |
| Ada Lovelace | Architecture | Blackwell 2.0 |
| AD102 | GPU Codename | GB206-300 |
| -Intel Core i9 14900Kor above | Recommended CPU | -Intel Core Ultra 7 255HXor above |
| 42169 | Steel Nomad Lite Score | 13896 |
| 36311 | Time Spy | 13563 |
| 187386 | Solar Bay | 59662 |
| 26128 | Port Royal | 8648 |
| 72387 | Fire Strike | 35860 |
| 85230 | Wild Life Extreme | 24526 |
| 195880 | Night Raid | 125756 |
| 2235 MHz | Base Clock | 890-1410 MHz (configurable) |
| 2520 MHz | Boost Clock | 1410-2520 MHz (configurable) |
| 16384 | Shading Units | 4608 |
| 512 | Texture Mapping Units (TMUs) | 144 |
| 176 | Render Output Units (ROPs) | 48 |
| 128 | Compute Units (Pipelines) | 36 |
| 512 | Tensor Cores | 144 |
| 128 | Ray-tracing Cores | 36 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 72MB shared | L2 Cache | 32MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |
Compare with something else?
Compare GeForce RTX 4090 with another →