Quick/Comparisons
41 | Gaming | 17 |
40 | Workstation | 24 |
38 | AI/ML |
Compare with something else?
Compare GeForce RTX 4070 with another →20 |
56 | Energy Efficiency | 100 |
43 | Quick Comparison Final Score | 33 |
| AD106 (GN21-X6) | Graphics Processor | Panther Lake |
| 4608 | Cores | — |
| 144 / 48 | TMUs / ROPs | — |
| 3587.08 | Blender GPU | — |
| 90 | Forza Horizon 5 | 38 |
| 132 | The Witcher 3 | 54 |
| 178 | Counter-Strike 2 | 61 |
| 112 | Far Cry 6 | 52 |
| 75 | Hogwarts Legacy | 24 |
| 104 | Call of Duty | 35 |
| 68 | Ghost of Tsushima | 27 |
| 87 | Cyberpunk 2077 | 33 |
| 139 | Tomb Raider | 50 |
| 109 | Average FPS | 42 |
| 131 | 1080p High | 54 |
| 109 | 1080p Ultra | 42 |
| 81 | 1440p Ultra | 29 |
| 46 | 4K Ultra | 15 |
| Low | Margin of Error | High |
| 115081 | GB6 Compute Score | 48273 |
| 158.5 img/sec | Background Blur | 117.4 img/sec |
| 116.5 img/sec | Face Detection | 60.6 img/sec |
| 4.35 Gpixels/sec | Horizon Detection | 1.39 Gpixels/sec |
| 5.67 Gpixels/sec | Edge Detection | 1.95 Gpixels/sec |
| 6.44 Gpixels/sec | Gaussian Blur | 2.45 Gpixels/sec |
| 1.36 Gpixels/sec | Feature Matching | 0.66 Gpixels/sec |
| 528.5 Gpixels/sec | Stereo Matching | 144.8 Gpixels/sec |
| 16372.3 FPS | Particle Physics | 7272.6 FPS |
| OpenCL | API | OpenCL |
| 19527 | G3D Mark Score | 8227 |
| 767 | G2D Mark | 800 |
| 178 FPS | DirectX 11 | 64 FPS |
| 85 FPS | DirectX 12 | 41 FPS |
| 8389 Ops/s | GPU Compute | 3327 Ops/s |
| 21186 | GB6 ML Single Precision | 10834 |
| 34404 | GB6 ML Half Precision | 20482 |
| 16559 | GB6 ML Quantized | 7349 |
| 8642 | Image Classification (SP) | 5389 |
| 20635 | Image Segmentation (HP) | 18899 |
| 26378 | Image Super Resolution (Q) | 8326 |
| 39776 | Face Detection (HP) | 25377 |
| 105983 | Pose Estimation (Q) | 16174 |
| 2995 | Text Classification (SP) | 1763 |
| 4507 | Machine Translation (HP) | 4140 |
| 10198 | Object Detection (SP) | 7138 |
| 41078 | Depth Estimation (Q) | 20034 |
| 190315 | Style Transfer (SP) | 84880 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 59-104 GPixel/s | Pixel Fill Rate | — |
| 177-313 GTexel/s | Texture Fill Rate | — |
| 11.3-20 TFLOPS | FLOPS (FP32) | — |
| GDDR6 | Memory Type | System Shared |
| 8 GB | Memory Size | — |
| 2000 MHz | Memory Clock | — |
| 16000 Mbps | Effective Memory Speed | — |
| 128-bit | Bus | — |
| No | ECC | No |
| 256 GB/s | Memory Bandwidth | — |
| PCIe 4.0 x16 | Interface | PCIe 5.0 |
| 35-140 W (configurable) | TGP | — |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 3 nm |
| 186 mm² | Die Size | — |
| 22.9 billion | Transistor Count | — |
| 123.12 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 |
| February 22, 2023 | Released | January 5, 2026 |
| Laptop | Case | Laptop |
| Gaming | Purpose | Generic |
| High-end | Segment | Mid-range |
| Ada Lovelace | Architecture | Celestial |
| AD106 (GN21-X6) | GPU Codename | Panther Lake |
| -Intel Core Ultra 7 155Hor above | Recommended CPU | -Intel Core Ultra 7 356Hor above |
| 12729 | Steel Nomad Lite Score | 5045 |
| 12518 | Time Spy | 6023 |
| 54826 | Solar Bay | — |
| 7403 | Port Royal | 3578 |
| 30044 | Fire Strike | 14710 |
| 23244 | Wild Life Extreme | — |
| 113720 | Night Raid | — |
| 735-2070 MHz (configurable) | Base Clock | — |
| 1230-2175 MHz (configurable) | Boost Clock | 2400 MHz |
| 4608 | Shading Units | — |
| 144 | Texture Mapping Units (TMUs) | — |
| 48 | Render Output Units (ROPs) | — |
| 36 | Compute Units (Pipelines) | 10 |
| 144 | Tensor Cores | — |
| 36 | Ray-tracing Cores | — |
| 128KB per cluster | L1 Cache | 64KB per cluster |
| 32MB shared | L2 Cache | 16MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |