NVIDIA has just announced in GTC 2025 its latest GPUS line for Workstations, RTX PRO 6000. Built on the innovative Blackwell architecture, the new graphic plate was created for high performance in professional applications with impressive specifications, including 24.064 NÚCLEOS CUDA, 96 GB of memory GDDR7, PCIE SUPPORT 5.0 and an energy consumption that can reach Up to 600W.
This new generation brings a performance leap to workstations, data centers and rendering serverstaking advantage of all the power of silicon GB – The same used in the expected GeForce RTX 5090but specifically configured for professional applications, although with reduced frequencies for thermal optimization and stability.
A new generation of professional GPUS
The RTX Pro 6000 family comes to succeed previous models, bringing significant advances in performance and efficiency.
The professional video card is designed to meet demanding computational needs, such as 3D modeling, scientific simulations and artificial intelligence. Its great differential is the optimization for heavy workflows for incomparable speed and accuracy.
Remembering that the older sister, RTX 6000 Ada Generation of 48 GB of RAM is now available today in Brazil for about $ 112,000 in partners like Dell. Therefore, it is expected that the pro version raises the values on levels yet but high.
Powerful technical specifications
With 24,064 CUDA nuclei and a 512 -bit memory interface, the RTX Pro 6000 offers a high bandwidth, allowing the agile processing of large data volumes. In addition, the 96 GB of GDDR6 with ECC memory provide essential reliability and stability for professional Mission Criticism environments.
The GPU follows the PCIE 4.0 x16 format to achieve compatibility with modern systems. NVIDIA also presented a variant in design blowerfocused on high density workstations, where active refrigeration is crucial.
Nvidia is also at the forefront in the professional and corporate segment with the arrival of RTX Pro 6000 Blackwell
The RTX Pro 6000 Blackwell has a maximum energy consumption estimated in 600Wa considerable value, but justified by the performance leap. Blackwell architecture introduces new thermal management techniques, making the plate more efficient compared to previous generations.
As we have seen, its great differential is in gross computational power and great memory capacity, essential for applications in scientific visualization, complex simulations and high level 3D rendering.
Technical Specifications
- Architecture: Blackwell (GB202)
- CUDA COLORS: 24.064
- Tensor Cores: 768
- RT Cores: 192 (4th generation, optimized for Ray Tracing)
- TMUS: 768
- ROPs: 192
- Memory: 96 GB GDDDR7 etc.
- Memory interface: 512-bit
- Energy consumption (TGP): 600 W
- Energy connector: 12V2×6
RTX PRO 6000 Blackwell Max-Q: design Compacto
In addition to the previous model, there is also the RTX PRO 6000 Blackwell Max-Qan optimized version of the line, focused on energy efficiency and better thermal use.
Inspired by the concept of “Max-Q”a term previously used by NVIDIA for high efficiency mobile GPUS, this variant was designed to offer balance between performance and energy consumption within the RTX Pro 6000 family.
RTX Pro 6000 Max-Q Highlights:
- Same powerful core of RTX Pro 6000but with a reduced TDP of 300W (half of the default version, which reaches 600W).
- 2 slots compact designallowing more efficient installation in multiple Workstations.
- Side Blower Cooling Systemoptimizing heat dissipation and avoiding overheating in environments with several installed plates.
The variant was developed for machines that require multiple GPUS operating togetherbut without compromising thermal efficiency or excessive energy consumption. Its format is suitable for visual effects, engineering and data science studioswhere the use of various graphics plates is essential to accelerate workflows.
Server version and ecosystem expansion
In addition to the traditional workstation version, Nvidia has also revealed a variant dedicated to servers, optimizing its use for data centers applications. This model features a differentiated design aimed at high performance cloud computing, artificial intelligence and remote rendering solutions.
Main Aracteristics
- For cloud rendering, AI inference and advanced modeling
- Optimized format for server racks
- Support for up to 8 gpus per server
Also read:
Availability
The GPUS of the line RTX PRO 6000 Blackwell will hit the global market on the following dates:
- April 2025: Versions for Desktops (RTX Pro 6000 and Max-Q)
- May 2025: distribution expansion to manufacturers
- Second semester of 2025: Launch of the version Server Edition
Despite the impressive specifications already disclosed, Nvidia has not yet revealed complete information about clocks, price and availability in Brazil.
Source: Nvidia via Videocardz and Tom’s Hardware
Join the Adrenaline offers group
Check out the main offers of hardware, components and other electronics we find over the internet. Video card, motherboard, RAM and everything you need to set up your PC. By participating in our group, you receive daily promotions and have early access to discount coupons.
Enter the group and enjoy the promotions
Source: https://www.adrenaline.com.br/nvidia/nvidia-rtx-pro-6000-blackwell-anunciada/
