- AMD can prioritize the price rather than specifications with Radeon Pro W9000’s limit of 32 GB
- AMD RADEON PRO W9000 targets
- With a 356 mm² die size, Radeon W9000 could still hit hard in reality tasks in the real world
AMD is ready to expand its workstation GPU -lineup with the upcoming release of the new Radeon Pro W9000 series, built on RDNA 4 architecture. This chip is aimed at professionals working in demanding fields such as video editing, 3D reproduction and AI development.
A leak from Hoang Anh Phu claims that the set -up will include a 32 GB model based on the new NAVI 48 XTW architecture.
AMD is assumed to hunt NVIDIA, but the RTX Pro 6000 Blackwell boasts a huge 96 GB frame buffer and 24,064 cuda kernels. This is far beyond the 32 GB configuration that AMD seems to offer, which suggests that the Radeon Pro W9000 may be targeted at professionals who do not need large data sets and instead appreciate performance per year. Dollars along with efficiency and affordable prices.
Does AMD change its approach?
As with all unconfirmed rumors, the details must be taken with a grain of salt – but if there is exactly, the leak suggests a shift in how AMD prioritizes the benefit in relation to costs. The previous generation of Radeon Pro W7900 contained 48 GB of memory, so 32 GB – probably GDDR6 – is a reduction.
While the capacity is lower, the Navi 48 XTW Die, which measures 356 mm², is still expected to deliver performance improvements across a wide range of tasks.
It is also likely to bring advantages to CAD workloads, CGI reproduction and real-time simulations, areas where the best laptops for video editing or photoshop also show gains.
The chip is expected to come in XL, XT and XTX variants, each set for different professional needs. The XTW model is assumed to have hardware improvements that are optimized for pro work, making it a strong opportunity for users who need a GPU for production of production quality.
Like Phu, a known hardware leakage, noticed, “It’s not as beef as the last generation, but pricing remains the key.”
A remaining question is software support. RDNA 4 does not yet have full integration with AMDS ROCM platform, which is critical for AI and machine learning developers.
Although the final specifications and an official name are still under packaging, all signs of a strategic disclosure point in line with Computex 2025 and AMD’s “Advancing AI” event in June 2025.
Via Toms hardware
