Quick/Comparisons
92 | Gaming | 16 |
91 | Workstation | 25 |
76 | AI/ML |
Compare with something else?
Compare GeForce RTX 4090 with another →22 |
43 | Energy Efficiency | 100 |
89 | Quick Comparison Final Score | 33 |
| AD102 | Graphics Processor | Panther Lake |
| 16384 | Cores | 1536 |
| 512 / 176 | TMUs / ROPs | 48 / 24 |
| 11688.48 | Blender GPU | 1268.79 |
| 203 | Forza Horizon 5 | 38 |
| 270 | The Witcher 3 | 55 |
| 245 | Counter-Strike 2 | 60 |
| 183 | Far Cry 6 | 49 |
| 165 | Hogwarts Legacy | 21 |
| 269 | Call of Duty | 32 |
| 150 | Ghost of Tsushima | 25 |
| 204 | Cyberpunk 2077 | 33 |
| 283 | Tomb Raider | 52 |
| 219 | Average FPS | 41 |
| 256 | 1080p High | 52 |
| 219 | 1080p Ultra | 41 |
| 176 | 1440p Ultra | 29 |
| 107 | 4K Ultra | 14 |
| Low | Margin of Error | Low |
| 2.38 Gpixels/sec | Feature Matching | — |
| 1950 Gpixels/sec | Stereo Matching | — |
| 48090.4 FPS | Particle Physics | — |
| OpenCL | API | — |
| 316301 | GB6 Compute Score | 64546 |
| 318.9 img/sec | Background Blur | — |
| 222.8 img/sec | Face Detection | — |
| 14.2 Gpixels/sec | Horizon Detection | — |
| 21 Gpixels/sec | Edge Detection | — |
| 23.9 Gpixels/sec | Gaussian Blur | — |
| 38071 | G3D Mark Score | 9029 |
| 1303 | G2D Mark | 786 |
| 324 FPS | DirectX 11 | 70 FPS |
| 151 FPS | DirectX 12 | 50 FPS |
| 26193 Ops/s | GPU Compute | 3702 Ops/s |
| 42962 | GB6 ML Single Precision | 11871 |
| 59636 | GB6 ML Half Precision | 20120 |
| 31670 | GB6 ML Quantized | 7879 |
| 15600 | Image Classification (SP) | 5487 |
| 36657 | Image Segmentation (HP) | 16864 |
| 48114 | Image Super Resolution (Q) | 8788 |
| 78111 | Face Detection (HP) | 25029 |
| 289639 | Pose Estimation (Q) | 18852 |
| 4074 | Text Classification (SP) | 2115 |
| 6349 | Machine Translation (HP) | 4266 |
| 21186 | Object Detection (SP) | 7449 |
| 75599 | Depth Estimation (Q) | 22207 |
| 601428 | Style Transfer (SP) | 95039 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 444 GPixel/s | Pixel Fill Rate | 60 GPixel/s |
| 1290 GTexel/s | Texture Fill Rate | 120 GTexel/s |
| 82.6 TFLOPS | FLOPS (FP32) | 7.7 TFLOPS |
| GDDR6X | Memory Type | System Shared |
| 24 GB | Memory Size | — |
| 2625 MHz | Memory Clock | — |
| 21000 Mbps | Effective Memory Speed | — |
| 384-bit | Bus | — |
| No | ECC | No |
| 1010 GB/s | Memory Bandwidth | — |
| PCIe 4.0 x16 | Interface | PCIe 5.0 |
| 450 W | TGP | — |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 3 nm |
| 609 mm² | Die Size | — |
| 76.3 billion | Transistor Count | — |
| 125.29 MTr/mm² | Transistor Density | — |
| 12 | DirectX | 12.1 |
| 1.3 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.9 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | No |
| 1.4a | DisplayPort | 2.1 |
| Nvidia | Vendor | Intel |
| Discrete | Build | Integrated |
| September 20, 2022 | Released | January 5, 2026 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Gaming |
| High-end | Segment | Mid-range |
| Ada Lovelace | Architecture | Celestial |
| AD102 | GPU Codename | Panther Lake |
| -Intel Core i9 14900Kor above | Recommended CPU | -Intel Core Ultra X9 388Hor above |
| 42169 | Steel Nomad Lite Score | 4242 |
| 36311 | Time Spy | 6595 |
| 187386 | Solar Bay | 26399 |
| 26128 | Port Royal | 3886 |
| 72387 | Fire Strike | 16770 |
| 85230 | Wild Life Extreme | 11707 |
| 195880 | Night Raid | 68498 |
| 2235 MHz | Base Clock | 300 MHz |
| 2520 MHz | Boost Clock | 2500 MHz |
| 16384 | Shading Units | 1536 |
| 512 | Texture Mapping Units (TMUs) | 48 |
| 176 | Render Output Units (ROPs) | 24 |
| 128 | Compute Units (Pipelines) | 12 |
| 512 | Tensor Cores | — |
| 128 | Ray-tracing Cores | — |
| 128KB per cluster | L1 Cache | 64KB per cluster |
| 72MB shared | L2 Cache | 16MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |