The GPU runs at a base clock of 2295 MHz and reaches a boost of 2550 MHz, with 10,240 shading units, 320 texture mapping units, and 112 render output units working together to produce a texture rate of 816 GTexels/s and a pixel fill rate of 285.6 GPixel/s. Floating-point performance is rated at 52.22 TFLOPS, reflecting a compute-capable specification suited to demanding workloads. GPU memory operates at 1438 MHz, and Double Precision Floating Point is supported, adding utility for tasks that require higher numerical precision beyond standard single-precision compute.
The card is equipped with 16GB of GDDR6X VRAM running across a 256-bit memory bus at an effective speed of 23,000 MHz, resulting in a maximum memory bandwidth of 736.3 GB/s. ECC memory support is included, which benefits users running compute workloads where data integrity is a priority alongside standard graphics tasks.
The card supports DirectX 12 Ultimate, OpenGL 4.6, and OpenCL 3, providing broad compatibility across modern graphics and compute workloads. Ray tracing, stereoscopic 3D, and multi-display technology are all supported, with up to four screens drivable simultaneously. DLSS is included for AI-based upscaling in compatible applications, while XeSS (XMX) is not available. Intel Resizable BAR is present, LHR is absent, and RGB lighting is integrated into the card's design.
The card provides four video outputs in total, comprising one HDMI 2.1a port and three DisplayPort outputs. There are no USB-C, DVI, or mini DisplayPort connections on this model.
Underpinned by the Ada Lovelace architecture, the GPU is built on a 5 nm process and integrates 45,900 million transistors, connecting to the host system via PCIe 4. The card carries a 320W TDP, which places meaningful demands on system power delivery and airflow; air-water cooling is not supported. Its physical footprint measures 300 mm in width and 120 mm in height, making case clearance an important consideration during system planning.
