Key Takeaways
- AMD has launched the Radeon AI PRO R9700, a new graphics card tailored for AI development on workstations.
- It features a substantial 32 GB of VRAM, a key advantage for memory-intensive AI tasks.
- AMD claims the R9700 can significantly outperform Nvidia’s RTX 5080, by up to 500% in specific AI workloads on Windows 11.
- The GPU is set for release in July, though pricing details are still under wraps.
AMD has unveiled a new graphics card, the Radeon AI PRO R9700, specifically designed for professionals working with artificial intelligence. This new GPU aims to meet the rising need for powerful hardware that can handle complex AI tasks directly on local workstations.
The R9700 shares some core technology with AMD’s gaming GPUs but truly distinguishes itself with a hefty 32 GB of video memory. This large memory capacity is specifically designed to accelerate AI tasks, which are known to be very memory-hungry.
AMD is making bold claims about the R9700’s capabilities. According to Neowin, AMD states the R9700 can “obliterate” Nvidia’s competing RTX 5080 in certain AI scenarios on Windows 11. They suggest it could be nearly 500% faster when working with large AI models, largely thanks to its increased memory.
The company also highlighted that combining four R9700 PRO GPUs can provide a staggering 128 GB of total VRAM for even more intensive AI work. Compared to AMD’s previous generation W7800, the new R9700 also promises up to double the performance in some cases, driven by improvements in its core architecture rather than just memory.
Pricing details haven’t been officially released yet. However, given its positioning against Nvidia’s RTX 5080, the cost might be in a similar range. Professionals looking to upgrade can expect the Radeon AI PRO R9700 to be available this July.
