NVIDIA RTX PRO 6000 Blackwell Server Edition

The NVIDIA RTX PRO 6000 Blackwell Server Edition is a passively cooled, dual-slot PCIe 5.0 GPU that brings the Blackwell architecture to standard rack servers. It carries 96 GB of GDDR7 with ECC on a 512-bit bus, 24,064 CUDA cores, 752 fifth-generation Tensor Cores and 188 fourth-generation RT Cores, with board power configurable up to 600 W to match the airflow of the host system. Unlike the HBM data-centre parts, it is a full graphics GPU, so the same card handles LLM inference, rendering, simulation and virtual workstations through NVIDIA vGPU, and Multi-Instance GPU can split it into up to four isolated partitions.

It is bought by enterprises that want one GPU type for mixed AI, visualisation and VDI workloads, by cloud providers offering Blackwell-based inference on PCIe infrastructure, and by broadcast, architecture and engineering firms running GPU farms. Most OEM 2U servers accept two to four cards, and 4U and 5U chassis from Supermicro, Dell, HPE and Lenovo take eight or more.

InTechnologies FZCO sources RTX PRO 6000 Server Edition cards and ready-built GPU servers for B2B customers in Dubai and the Middle East. The hardware is supplied under contract and invoice from our licensed partners, with InTechnologies responsible for sourcing, consolidation in Dubai and DDP or DDU delivery. Professional Blackwell GPUs are subject to US export regulations, and we carry out end-user verification before accepting an order.