Nvidia has licensed mental property from inferencing chip designer Groq, and employed away a few of its senior executives, however stopped in need of an outright acquisition.
“We’ve taken a non-exclusive license to Groq’s IP and have employed engineering expertise from Groq’s crew to hitch us in our mission to supply world-leading accelerated computing expertise,” an Nvidia spokesman stated Tuesday, by way of e mail. However, he stated, “We haven’t acquired Groq.”
Groq designs and sells chips optimized for AI inferencing. These chips, which Groq calls language processing items (LPUs), are lower-powered, lower-priced units than the GPUs Nvidia designs and sells, which nowadays are primarily used for coaching AI fashions. Because the AI market matures, and utilization shifts from the creation of AI instruments to their use, demand for units optimized for inferencing is more likely to develop.