Nscale Tests NVIDIA DSX Solution, Boosting Inference Throughput by 49%
15 hour ago / Read about 0 minute
Author:小编   

At its AI factory located in Iceland, Nscale conducted an evaluation on four NVIDIA GB300 NVL72 racks, employing the DSX MaxLPS system. The test outcomes revealed a significant 49% surge in inference throughput, all while adhering to a fixed power budget of 264 kilowatts. Additionally, there was a notable 37% expansion in the manageable GPU capacity.