Nvidia to License Groq’s AI Chip Technology in $20 Billion Agreement

Nvidia has entered a non-exclusive licensing agreement with AI chip startup Groq, a deal valued at approximately $20 billion. The agreement covers Groq’s inference technology and includes the hiring of several key executives, including founder and CEO Jonathan Ross and president Sunny Madra. Groq’s chief financial officer, Simon Edwards, will assume the role of CEO following the transaction.

The arrangement allows Nvidia to integrate Groq’s low-latency processors into its AI factory architecture to support a wider range of real-time and inference workloads. Despite the scale of the deal, Groq will continue operating as an independent company, and its GroqCloud service will remain unaffected.

Groq’s flagship product, the LPU (Language Processing Unit), is designed for AI inference tasks and is claimed to deliver up to ten times greater energy efficiency than conventional graphics processors. The chip’s architecture includes on-chip SRAM and a deterministic processing design that minimizes delays in computation. Groq also connects its LPUs using a proprietary interconnect technology called RealScale, developed to improve synchronization across AI servers.

The reported $20 billion price represents a significant premium over Groq’s $6.9 billion valuation from its September funding round. The transaction marks Nvidia’s largest technology licensing deal to date and adds to a series of similar agreements aimed at expanding its AI hardware capabilities.