Nvidia has expanded its professional workstation lineup with the RTX Pro 5500 Blackwell Workstation Edition. The new model is positioned between the flagship RTX Pro 6000 and the entry-level RTX Pro 5000, offering higher memory capacity and power limits than the latter.
The card features a GPU with 21,760 CUDA cores and 84 GB of GDDR7 ECC memory on a 416-bit bus. It delivers a memory bandwidth of 1,398 GB/s and has a maximum power consumption of 600 W. The hardware utilizes a PCIe 5.0 x16 interface and provides four DisplayPort 2.1b outputs.
Equipped with three ninth-generation NVENC encoders and three sixth-generation NVDEC decoders, the card supports Multi-Instance GPU technology. This allows the unit to operate as a single 84 GB instance or two isolated 42 GB instances. It is available with active air cooling or the RXM liquid cooling system.
Compared to the flagship RTX Pro 6000, the new model has 9.6% fewer CUDA cores and 12.5% less memory. While the RTX Pro 6000 reaches a bandwidth of 1,792 GB/s, the RTX Pro 5500 is limited to 1,398 GB/s. The RTX Pro 5000 remains the lower-tier option with a 300 W power limit.
The RTX Pro 5500 is listed for markets including the US, UK, India, and Japan, with PNY also publishing product specifications. Nvidia has not yet disclosed the pricing for the new workstation card, noting that availability is expected soon and current specifications remain preliminary.
