AMD is reportedly going to launch a special Radeon RX 9070 XT GPU featuring up to 32 GB VRAM and even though it might seem like a PRO GPU, it's supposedly for gaming.
AMD Radeon RX 9070 XT 32 GB Rumored to be in Works: Might Release in the First Half of 2025
While AMD hasn't even launched a single RDNA 4 GPU, we are hearing reports about more SKUs reportedly coming this year. AMD's RDNA 4 lineup, i.e., RX 9000 series as we know, features Navi 48 and Navi 44-based GPUs and the first ones to launch will be the RX 9070 XT and RX 9070 in March.
As far as their specifications go, these GPUs will be shipped with 16 GB of GDDR6 VRAM, but some rumors suggest AMD could be working on releasing another RX 9070 series GPU but with twice the VRAM capacity. Chiphell Forum member 'zhangzhonghao' indicated that AMD is planning another RX 9070/XT version with a larger video memory of up to 32 GB, double that of the standard variants that feature 16 GB memory.
Furthermore, he clarifies in the original post that it may get a release date this year, probably the first half. In a recent post, he further stated that its launch may happen before the end of Q2. As you may be wondering if this is a PRO card for professional workloads, then the leaker suggests otherwise. The RX 9070 or RX 9070 XT with 32 GB is supposedly being released in the gaming lineup but is aimed at AI workloads that require higher VRAM, but it can still game just like the 16 GB editions.
That said, if the GPU gets 32 GB VRAM capacity, AMD will have to deploy 16x 2 GB GDDR6 memory modules, which isn't going to be possible on just the front side. AMD will have to deploy half of them at the back to achieve a 32 GB VRAM configuration. Currently, the 4 GB GDDR6 memory modules don't exist and this will for sure increase the cost of the GPU noticeably due to newer challenges.
The bus memory width will remain the same at 256-bit and, since the memory speed of 20 Gbps will be retained as well, the total memory bandwidth will be unaffected. This will be AMD's first 32 GB gaming GPU, which on paper looks amazing, but in reality, it won't be powerful enough to leverage the potential of such a huge VRAM size in gaming. However, in other memory-intensive tasks such as AI (large LLM models), this should make things quicker.
About the author: Sarfraz Khan is a hardware reporter with a focus on PC components and the builder community. With years of experience writing about PC hardware and laptops, his work has been featured on several reputable technology publications. Sarfraz's hands-on experience is demonstrated through his first-person accounts of using and comparing different hardware configurations, providing practical and relatable insights for everyday users. His technical analysis is respected by peers in the enthusiast community and has been cited by specialized hardware sites such as Germany's Igor's Lab.
Follow Wccftech on Google to get more of our news coverage in your feeds.
