6 days ago
48
Groq just secured $650 million in growth funding, announced on June 22, 2026. That’s a hefty check for a company that recently watched its founder, CEO, and president walk out the door to join Nvidia.
The round was led by Disruptive, whose founder Alex Davis also chairs Groq’s board, along with Infinitum. Existing investors opted to reinvest as well.
The Nvidia deal that changed everything
In December 2025, Nvidia struck a non-exclusive licensing agreement with Groq valued at roughly $20 billion, targeting Groq’s inference technology for its own platforms. As part of the arrangement, Nvidia hired several of Groq’s most important people. Founder and CEO Jonathan Ross left. So did president Sunny Madra.
Groq’s previous post-money valuation stood at $6.9 billion following a $750 million round in September 2025. The current valuation has not been disclosed, though earlier reporting pegged it around $3.5 billion.
Rebuilding from the inside out
Groq has essentially hit the reset button on its leadership team. Alan Rice, who previously worked at both xAI and Meta, has joined as chief operating officer. Sinclair Schuller stepped in as CTO, and Rakesh Malhotra took the chief product officer role.
Groq currently operates 13 data centers worldwide, serves more than five million developers, and processes trillions of AI tokens on a weekly basis.
The company’s Language Processing Unit, or LPU, was originally designed as a specialized chip for inference workloads. Groq’s technology earned enough attention that Nvidia decided to license it rather than compete with it. That licensing deal will result in Nvidia deploying what it calls the LPX platform, which incorporates Groq’s inference technology. Groq itself plans to deploy Nvidia’s LPX platform within its own data centers as part of this expansion.
Where the money goes
The $650 million is earmarked for scaling Groq’s infrastructure significantly. The company is targeting 200 megawatts of power capacity by the end of 2027. Groq is positioning itself as an infrastructure provider, going all-in on operating its own inference cloud rather than competing in the hardware market.
Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.
English (US) ·