The GPU operates at a base clock of 2340 MHz and boosts up to 2610 MHz, supported by 8448 shading units, 264 texture mapping units, and 96 render output units that together deliver a texture rate of 689 GTexels/s and a pixel fill rate of 250.6 GPixel/s. Floating-point performance is rated at 44.1 TFLOPS, with GPU memory running at 1313 MHz. Double Precision Floating Point is also supported, adding utility for compute workloads that depend on higher numerical precision beyond standard single-precision tasks.
The card is fitted with 16GB of GDDR6X VRAM running across a 256-bit memory bus at an effective speed of 21,000 MHz, producing a maximum memory bandwidth of 672.3 GB/s. ECC memory support is included, which helps maintain data integrity for users running compute or professional GPU workloads alongside standard graphics tasks.
The card supports DirectX 12 Ultimate, OpenGL 4.6, and OpenCL 3, providing solid coverage across modern graphics and compute environments. Ray tracing, stereoscopic 3D, and multi-display technology are all present, with up to four screens supported simultaneously. DLSS is available for AI-based upscaling in compatible titles and applications, while XeSS (XMX) is not supported. Intel Resizable BAR is included, LHR is absent, and RGB lighting is integrated into the card's design.
The card offers four video outputs in total, made up of one HDMI 2.1a port and three DisplayPort outputs. USB-C, DVI, and mini DisplayPort connections are not included on this model.
Built on the Ada Lovelace architecture and produced using a 5 nm process, the GPU packs 45,900 million transistors and interfaces with the host system via PCIe 4. The card's 285W TDP means system builders need to account for adequate power delivery and airflow, and air-water cooling is not supported. At 300 mm wide and 120 mm tall, the card has a relatively manageable footprint that should fit comfortably in most standard full-size enclosures.
