- AMD Radeon AI PRO R9700S delivers 47.8 TFLOPS peak FP32 compute energy
- AMD equips the R9700S with 32GB GDDR6 reminiscence for AI workloads
- PCIe 5.0 x16 assist allows high-bandwidth communication for AI purposes
AMD has launched the Radeon AI PRO R9700S, a passively cooled workstation GPU that joins the R9000 collection and guarantees enterprise-level AI efficiency.
After the launch of this collection earlier this yr, a number of main manufacturers confirmed merchandise that includes R9700 playing cards, indicating sturdy trade adoption.
The R9700S retains the identical Navi 48 RDNA4 configuration as the usual R9700, providing 64 compute models and 4096 stream processors.
Passive cooling for high-density AI workloads
Its enhance clock reaches as much as 2920MHz and delivers a peak FP32 throughput of 47.8 TFLOPS, which makes it a succesful possibility for enterprise AI workloads.
Designed for dense racks and multi-GPU setups, the R9700S makes use of a 32GB GDDR6 reminiscence buffer on a 256-bit bus with 64MB Infinity Cache.
PCIe 5.0 x16 assist ensures high-bandwidth communication with appropriate workstation platforms.
The “S” designation within the R9700S signifies a silent design, which replaces the blower-style cooler of earlier fashions.
This method makes use of system airflow quite than onboard followers, making the GPU appropriate for compact racks the place a number of playing cards function intently.
Regardless of the passive cooling, the cardboard retains a 300W TDP and attracts energy by a single 12V-2×6 connector.
This method displays AMD’s concentrate on sustaining compute efficiency with out including lively elements.
The corporate additionally launched the R9600D, which scales all the way down to 48 compute models and 3072 stream processors whereas protecting an identical reminiscence.
Nonetheless, the R9700S affords greater throughput for giant AI fashions.
Each playing cards assist Linux ECC reminiscence choices and are totally appropriate with AMD’s Software program PRO Version and ROCm integration.
The R9700S is optimized for duties equivalent to generative AI inference and coaching of huge language fashions.
This lets enterprises offload demanding workloads from CPU cores to GPU accelerators.
Early adoption of the R9700 has appeared in prebuilt methods like Elsa’s Veluga-D A70S G6 workstation.
It pairs the GPU with an AMD Ryzen 7 9700X CPU, 64GB of DDR5 RAM, and PCIe growth slots for added accelerators.
With these chips, AMD affords enterprise-ready options for AI computation whereas protecting a concentrate on reminiscence bandwidth, energy supply, and silent operation.
That stated, the excessive uncooked efficiency of this chip raises questions on sustained thermals in dense deployments.
Due to this fact, enterprises that deploy a number of models should think about airflow and system-level warmth administration to take care of stability below peak AI workloads.
Through Videocardz
Observe TechRadar on Google Information and add us as a most well-liked supply to get our knowledgeable information, critiques, and opinion in your feeds. Be certain to click on the Observe button!
And naturally you may as well observe TechRadar on TikTok for information, critiques, unboxings in video type, and get common updates from us on WhatsApp too.
