Chip giant NVIDIA (NVDA) has agreed to acquire key core assets of the high-performance artificial intelligence acceleration chip startup Groq for $20 billion in cash. This transaction is not only NVIDIA’s largest deal ever, far surpassing its previous record set in 2019 with the approximately $7 billion acquisition of Israeli chip design company Mellanox, but also marks a further deepening of its strategic layout in the artificial intelligence chip sector. As of the end of October this year, NVIDIA’s cash and short-term investments amounted to a substantial $60.6 billion, providing ample funding for this significant investment.
The structure of this transaction is notably distinctive. According to Alex Davis, CEO of Disruptive, a firm involved in Groq’s financing, NVIDIA will acquire nearly all of Groq’s assets. However, Groq’s cloud computing business, GroqCloud, is excluded from the deal and will continue to operate independently. Concurrently, Groq has entered into a non-exclusive licensing agreement with NVIDIA, granting the latter rights to its AI inference technology. Following the completion of the transaction, Groq’s founder and CEO Jonathan Ross, President Sunny Madra, and several other key executives will join NVIDIA to assist in advancing related technologies. Groq as a company will continue to exist as an “independently operated enterprise,” with former Chief Financial Officer Simon Edwards assuming the role of the new CEO.
NVIDIA has maintained a low profile regarding the specifics of the transaction, but its strategic intent is clear. In an internal email, NVIDIA CEO Jensen Huang indicated plans to integrate Groq’s low-latency processors into NVIDIA’s AI Factory architecture to address a broader range of AI inference and real-time workloads. He emphasized that this is not an outright acquisition of Groq but rather an effort to bring in key talent and license its core intellectual property. This structural transaction, focused on talent and technology, is not unprecedented for NVIDIA. In September of this year, the company employed a similar approach, investing over $900 million to onboard the CEO and team of AI hardware startup Enfabrica. Tech giants such as Meta, Google, and Microsoft also frequently adopt similar models to compete for top-tier AI talent.
Despite recent pressures on the artificial intelligence-related sector, some analysts maintain an optimistic outlook for industry leaders. In a report, Cantor Fitzgerald analyst C.J. Muse noted that NVIDIA and Broadcom appear “well-positioned to outperform the market in 2026.” He highlighted that a new cycle based on the Blackwell architecture is about to commence and that NVIDIA’s current stock valuation is attractive. Consequently, the firm has confidence in listing NVIDIA and Broadcom as top picks for 2026 and advises investors to position themselves early.
