For universal AI and Visual computing, the NVDIA RTX PRO 6000 Blackwell Server Edition offers outstanding performance for data center applications. It features 96GB of GDDR7 memory to tackle complex applications like AI, physical AI, plus scientific computing, 3D graphics and video. This GPU offers 5x the performance compared to the previous generation for large language models thanks to 5th generation tensor cores.
In addition to 752 5th generation tensor cores, the NVDIA RTX PRO 6000 Blackwell Server Edition also delivers 188 fourth-generation ray-tracing (RT) cores, next-gen video engines, and 96GB of GDDR7 memory. With DLSS 4, it can provide exceptional image quality and smoother, faster frame rates for life-like simulations. With Universal MIG (multi-Instance GPU) functionality, a single GPU can be divided up into up to 4x fully isolated instances of 24GB each allowing for the simultaneous execution of diverse workloads. Along with 24,064 CUDA parallel processing cores, memory bandwidth is listed at 1597GB/s with a 512-bit memory interface.
The card offers 4x DisplayPort 2.1 ports for support of up to 4x monitors and occupies a PCIe 5.0 x16 slot. With DisplayPort 2.1, it can power high-resolution displays up to 8Kat 240Hz and 16K at 60Hz. With high-definition resolution, it delivers superior color accuracy and color depth. Depending on the host server for air flow, the unit is passively cooled and requires a 16-pin power connector. At full power, it consumes up to 600W (configurable) with a recommended host PSU of up to 1000W.
If you know what you want but can't find the exact configuration you're looking for, have one of our knowledgeable sales staff contact you. Give us a list of the components you would like to incorporate into the system, and the quantities, if more than one. We will get back to you immediately with an official quote.