Nvidia’s planned US$20 billion strategic deal with Groq is built on a simple logic: as compute gets cheaper, demand keeps expanding. In a recent interview, Groq co-founder and CEO Jonathan Ross explained why Nvidia is expected to combine Groq’s LPU with its latest Vera Rubin platform and how GPU and LPU can work as complementary engines in LLM inference.
Groq CEO sees GPU and LPU as complementary as AI compute demand grows
22
Jun