- Details of Nvidia’s Fastest Ever WorkStation -Video Morts have been viewed online
- There are RTX Pro 6000 Blackwell and RTX Pro 6000 x Blackwell -Taste
- The non-X version has 24,064 Cuda-Kernes and 96 GB GDDR7 ECC memory
Online leaks seem to reveal details of NVIDIA’s upcoming RTX Pro 6000 Blackwell GPU, which promises to be the company’s most powerful workstation card to date.
Information discovered on the LEADTEK site and supported by NBD sending data points to the map rocking 24,064 CUDA -Kerner and 96 GB GDDR7 memory.
These specifications are significantly higher than both the current RTX 6000 ADA, which was launched over two years ago, and the expected RTX 5090 gaming GPU, which reportedly has 2,304 fewer cores.
NBD’s shipping data shows two versions of WorkStation GPU – RTX Pro 6000 Blackwell and RTX Pro 6000 x Blackwell – Sent to India for “Test Casse”. We are not sure what x means or how different it will be from the second version. However, it is noteworthy that Nvidia seems to identify its new cards with a pro -label.
The information on one of these maps (the non-X version) was placed on X by Harukaze, which shows that GPU is running 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 crucial.
Power Hungry Beast
The power requirements are especially high for the new GPU with a 600W total graphics power delivered through a 16-pole PCIE 5.0 connector. This puts it among the most power -hungry GPUs in Nvidia’s lineup – over twice required by the RTX 6000 ADA – but it should not come as too much of a surprise. Unlike previous workstation maps, this one is likely to skip the traditional fan-style cools over and take its design signals from the RTX 5090 with a double flow refrigeration system.
Writing about the leak, Toms Hardware Remarks, “Workstation GPUs under Nvidia’s RTX lineup are tailored to professional applications. In most scenarios, however, 96 GB VRAM is excessive unless you are looking for training or inference AI locally. Although you would typically not use these GPUs for games, the added memory can be useful in tasks involved in game development, content creation, province and computer -supported design. “
GPU is expected to be officially introduced at NVIDIA’s GPU Technology Conference (GTC) in March 2025. Pricing is not confirmed, but it is likely to fall in the same advanced range as its predecessor, which may be for between $ 6,000 and $ 8,000.
