Quick/Comparisons
57 | Gaming | 45 |
49 | Workstation | 43 |
56 | AI/ML |
Compare with something else?
Compare Radeon RX 6900 with another →46 |
36 | Energy Efficiency | 55 |
54 | Quick Comparison Final Score | 44 |
| Navi 21 | Graphics Processor | Navi 44 |
| 5120 | Cores | 2048 |
| 320 / 128 | TMUs / ROPs | 128 / 64 |
| 2484.02 | Blender GPU | 1601.83 |
| 139 | Forza Horizon 5 | 112 |
| 185 | The Witcher 3 | 148 |
| 208 | Counter-Strike 2 | 185 |
| 152 | Far Cry 6 | 126 |
| 103 | Hogwarts Legacy | 88 |
| 179 | Call of Duty | 133 |
| 112 | Ghost of Tsushima | 90 |
| 127 | Cyberpunk 2077 | 97 |
| 208 | Tomb Raider | 162 |
| 157 | Average FPS | 127 |
| 188 | 1080p High | 155 |
| 157 | 1080p Ultra | 127 |
| 121 | 1440p Ultra | 94 |
| 71 | 4K Ultra | 54 |
| Low | Margin of Error | Low |
| 124121 | GB6 Compute Score | 107640 |
| 231.6 img/sec | Background Blur | 269.5 img/sec |
| 152.5 img/sec | Face Detection | 151 img/sec |
| 4.26 Gpixels/sec | Horizon Detection | 3.6 Gpixels/sec |
| 5.75 Gpixels/sec | Edge Detection | 4.67 Gpixels/sec |
| 5.87 Gpixels/sec | Gaussian Blur | 5.27 Gpixels/sec |
| 1.12 Gpixels/sec | Feature Matching | 1.08 Gpixels/sec |
| 618.7 Gpixels/sec | Stereo Matching | 395.3 Gpixels/sec |
| 17973.9 FPS | Particle Physics | 13165.7 FPS |
| OpenCL | API | OpenCL |
| 26671 | G3D Mark Score | 20129 |
| 1062 | G2D Mark | 1178 |
| 279 FPS | DirectX 11 | 185 FPS |
| 111 FPS | DirectX 12 | 57 FPS |
| 14356 Ops/s | GPU Compute | 11202 Ops/s |
| 33651 | GB6 ML Single Precision | 25924 |
| 36994 | GB6 ML Half Precision | 40429 |
| 24546 | GB6 ML Quantized | 22275 |
| 13868 | Image Classification (SP) | 12380 |
| 17467 | Image Segmentation (HP) | 15402 |
| 32396 | Image Super Resolution (Q) | 32024 |
| 47808 | Face Detection (HP) | 50457 |
| 178873 | Pose Estimation (Q) | 113796 |
| 4358 | Text Classification (SP) | 4460 |
| 7910 | Machine Translation (HP) | 7481 |
| 15394 | Object Detection (SP) | 11256 |
| 57351 | Depth Estimation (Q) | 53978 |
| 372610 | Style Transfer (SP) | 235750 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 288 GPixel/s | Pixel Fill Rate | 200 GPixel/s |
| 720 GTexel/s | Texture Fill Rate | 401 GTexel/s |
| 23 TFLOPS | FLOPS (FP32) | 25.6 TFLOPS |
| GDDR6 | Memory Type | GDDR6 |
| 16 GB | Memory Size | 16 GB |
| 2000 MHz | Memory Clock | 2518 MHz |
| 16000 Mbps | Effective Memory Speed | 20100 Mbps |
| 256-bit | Bus | 128-bit |
| No | ECC | No |
| 512 GB/s | Memory Bandwidth | 320 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 300 W | TGP | 160 W |
| TSMC | Manufacturing | TSMC |
| 7 nm | Fabrication Process | 4 nm |
| 520 mm² | Die Size | 199 mm² |
| 26 billion | Transistor Count | 29.7 billion |
| 50 MTr/mm² | Transistor Density | 149.25 MTr/mm² |
| 12 | DirectX | 12.2 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 2.1 | OpenCL | 2.2 |
| — | CUDA | No |
| Yes | Ray Tracing | Yes |
| No | DLSS | No |
| 1.4a | DisplayPort | 2.1a |
| Amd | Vendor | Amd |
| Discrete | Build | Discrete |
| December 8, 2020 | Released | June 4, 2025 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| High-end | Segment | Mid-range |
| RDNA 2.0 | Architecture | RDNA 4.0 |
| Navi 21 | GPU Codename | Navi 44 |
| -Intel Core i7 12700Kor above | Recommended CPU | -Intel Core Ultra 5 245Kor above |
| 20125 | Steel Nomad Lite Score | 13915 |
| 20809 | Time Spy | 16405 |
| 86980 | Solar Bay | 65446 |
| 10539 | Port Royal | 9751 |
| 57319 | Fire Strike | 38969 |
| 39341 | Wild Life Extreme | 27779 |
| 159586 | Night Raid | 147967 |
| 1825 MHz | Base Clock | 1700 MHz |
| 2250 MHz | Boost Clock | 3130 MHz |
| 5120 | Shading Units | 2048 |
| 320 | Texture Mapping Units (TMUs) | 128 |
| 128 | Render Output Units (ROPs) | 64 |
| — | Compute Units (Pipelines) | 32 |
| — | Tensor Cores | 64 |
| 80 | Ray-tracing Cores | 32 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 4MB shared | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 4 IPC |