Nvidia Corp. and AI accelerator startup Groq reached a $20 billion [1] agreement to integrate high-speed inference technology into a new chip platform.
The deal represents the largest transaction in Nvidia's three-decade history [2]. By leveraging Groq's specialized hardware, Nvidia aims to strengthen its dominance in the AI market and accelerate the speed at which AI models process data.
Reports indicate the agreement was finalized on Dec. 24, 2025 [3], though the details only became public in March 2026 [4]. The timing of the deal coincided with a period of intense competition in the AI hardware sector, a race to reduce latency in large language models.
There is conflicting information regarding the exact nature of the transaction. Some reports describe the arrangement as an all-cash acquisition of Groq by Nvidia [1]. Other sources said the deal is a licensing agreement and that Nvidia did not acquire the company [5].
Despite the lack of consensus on the corporate structure, the financial scale remains consistent at $20 billion [1]. Groq co-founder Ross said the deal was a whirlwind in an interview with Forbes earlier this year [4].
The agreement has already attracted the attention of U.S. lawmakers [6]. Some members of Congress have raised questions regarding the deal's impact on market competition, and the concentration of AI power within a single entity [7].
Nvidia is betting that Groq's technology will provide a critical edge for its next generation of AI chips [8]. The integration is expected to power a newly announced platform designed for enterprise-level AI deployment.
“The deal represents the largest transaction in Nvidia's three-decade history.”
This transaction signals a shift in the AI hardware race from raw computing power toward inference efficiency. By absorbing or licensing Groq's technology, Nvidia is attempting to solve the 'latency bottleneck' that slows down real-time AI responses. However, the ambiguity over whether this is a licensing deal or a full acquisition—combined with congressional scrutiny—suggests that antitrust regulators may view the move as an attempt to stifle emerging competition in the AI accelerator space.
