- Details of the video station’s video card from NVIDIA were identified online
- There is RTX Pro 6000 Blackwell and RTX Pro 6000 x Blackwell Flavors
- The non x version has 24,064 cuda cores and 96 GB of ECC GDDR7 memory
Online leaks seem to reveal the details of the next GPU RTX Pro 6000 Blackwell, which promises to be the most powerful work card of the company to date.
Information identified on the Leadtek website and supported by NBD Data Shipping, indicate the card that shakes 24,064 cuda cores and 96 GB of GDDR7 memory.
These specifications are significantly higher than the current RTX 6000 ADA, which was launched more than two years ago, and the expected RTX 5090 game GPU, which would include 2,304 less cores.
NBD shipping data show two versions of the GPU of the workstation – the RTX Pro 6000 Blackwell and the RTX Pro 6000 x Blackwell – sent to India for test purposes. We don’t really know what X means, or how different it will be from the other version. However, it should be noted that Nvidia seems to identify her new cards with a professional label.
Information on one of these cards (the Non X version) was published on X by Harukaze, showing that the GPU runs on a 512 -bit memory interface, and GDDR7 supports ECC (error correction), making it a good choice for servers and systems where data accuracy is essential.
Power hungry beast
The power requirements are notably high for the new GPU, with a total graphical power of 600 W delivered via a PCIe 5.0 to 16 pin connector. This places it among the most eager for NVIDIA GPUs – more than twice that required by RTX 6000 ADA – but that should not be too surprising. Unlike previous workstation cards, it will probably jump the traditional blowing style cooler and take its RTX 5090 design signals, with a double flow cooling system.
Write on the leak, Tom material Note: “The GPUs of the workstation as part of the NVIDIA RTX range are tailor -made for professional applications. However, in most scenarios, 96 GB of VRAM are exaggerated unless you are looking for training or inference AI locally. Although you do not generally use these GPUs for games, the added memory can be useful in the tasks involved in the development of games, content creation, service and computer -assisted design. »»
The GPU should be officially introduced at the NVIDIA GPU Technology (GTC) conference in March 2025. The price has not been confirmed, but it will probably fall into the same high -end range as its predecessor, which can be made between $ 6,000 and $ 8,000.
