Nvidia Makes A $20 Billion Push Toward Buying AI Chip Startup Groq

Nvidia has entered a new phase in its relationship with Groq by licensing the startup’s technology while also welcoming several senior leaders from the company. Groq said in a blog post that Nvidia has agreed to a “non-exclusive” license for its technology, and that founder Jonathan Ross, Groq President Sunny Madra and members of its engineering team will join Nvidia.

Groq added that it will continue operating as an independent company, with Simon Edwards taking over as CEO and its cloud business remaining active. A person close to Nvidia also confirmed the licensing agreement.

ALSO READ: iPhone SE To MacBook Air, Products That Apple Phased Out In 2025

Financial details were not disclosed by Groq, but CNBC reported that Nvidia had agreed to acquire Groq for $20 billion in cash. Neither Nvidia nor Groq commented on that report, and Groq reiterated in its post that it will keep running independently while the licensing and leadership transition moves ahead.

Groq has built a distinct presence in the AI inference space, where already trained models respond to real-time user queries. This is an area where competition is far broader than training, with rivals and newer players looking to gain relevance as the market shifts toward deployment and performance at scale.

In recent years, major technology firms have pursued deals that secure technology access and top leadership without full acquisitions, drawing close attention from regulators. Analysts have begun evaluating how this latest move fits into that pattern.

“Antitrust would seem to be the primary risk here, though structuring the deal as a non-exclusive license may keep the fiction of competition alive (even as Groq’s leadership and, we would presume, technical talent move over to Nvidia),” Bernstein analyst Stacy Rasgon wrote. The note also said that Nvidia CEO Jensen Huang’s “relationship with the Trump administration appears among the strongest of the key US tech companies.”

Groq more than doubled its valuation to $6.9 billion after a major funding round last year, and its SRAM-based on-chip memory approach helps accelerate AI interactions while limiting the size of models it can serve. Its closest rival in this design direction is Cerebras Systems, with both companies signing large deals in the Middle East as momentum builds around inference-focused computing.

ALSO READ: Gaming Could Cost A Lot More As Demand For RAM Surges: Why Are The Prices Rising?

Against this backdrop, Nvidia is positioning itself for the next shift in AI markets, while Groq signals independence even as talent, technology access and reported price speculation keep the conversation open-ended.