Groq Raised $350 Million to Become Nvidia's Landlord Instead of Its Rival

Groq raised $650 million in a round led by Disruptive and Infinitum to scale its AI inference cloud, following a December 2025 deal in which Nvidia licensed Groq's technology for about $20 billion and hired founder Jonathan Ross and president Sunny Madra. The company, now led by CEO Adam Winter and CFO Matt Eng, operates 13 data centers, serves over five million developers, and aims to reach 200 megawatts of capacity by the end of 2027, but it did not disclose a new valuation after being valued at $6.9 billion in September 2025.

Groq's latest raise is not a victory lap over Nvidia. It's the clearest sign that Groq now has to make money inside Nvidia's world. Groq raised $650 million in June to scale its AI inference cloud, and the useful part of the story is not the size of the check. The useful part is what Groq has become after Nvidia licensed its technology and hired away its founder, president and much of the team behind the chips that made the company famous. The challenger is still alive. It just looks less like a chip rebel now. According to TechCrunch, the new round was led by Disruptive, the Dallas investment firm founded by Alex Davis, and Infinitum, a Fort Lauderdale hedge fund. Groq did not disclose a new valuation. That detail matters because the company was last valued at $6.9 billion after a $750 million round in September 2025, also reported by TechCrunch. When a company raises fresh capital after a deal as strange as Groq's December agreement with Nvidia, the missing valuation is not a footnote. It's part of the story. The December deal was not a normal acquisition. Axios reported at the time that Nvidia agreed to a non-exclusive license for Groq's technology in a transaction widely pegged at about $20 billion, while Groq founder Jonathan Ross and president Sunny Madra moved to Nvidia. Groq stayed independent. Simon Edwards, then CFO, was named CEO after Ross left, according to Axios. By June, Groq's own announcement listed Adam Winter as CEO and Matt Eng as CFO. That's a lot of motion for a company that once had a very simple pitch: its language processing units, or LPUs, could run AI inference faster and more efficiently than the GPUs everybody was fighting to buy from Nvidia. Groq built its name on that claim. Now Nvidia has access to the technology, and Groq has to prove the cloud business around it can stand on its own. Groq raises $650 million for its neocloud second act after selling its soul to Nvidia for $20 billion https://startupfortune.com/groq-raises-650-million-for-its-neocloud-second-act-after-selling-its-soul-to-nvidia-for-20-billion/ Groq is raising $650 million to fund its pivot from AI chip maker to inference neocloud provider, following a $20 billion LPU licensing deal with Nvidia that paid out shareholders and handed the founding team to its former competitor. Existing investors Disruptive and Infinitum are backstopping the full round. The raise signals durable investor... - Groq raises 650 million for AI inference neocloud https://startupfortune.com/groq-raises-650-million-for-its-neocloud-second-act-after-selling-its-soul-to-nvidia-for-20-billion/ - AI chip startup licensing technology to Nvidia deal https://startupfortune.com/groq-raises-650-million-for-its-neocloud-second-act-after-selling-its-soul-to-nvidia-for-20-billion/ Here's the thing: that is still a real business. Groq says it operates 13 data centers across North America, Europe, the Middle East and Asia Pacific, serves more than five million developers, and processes trillions of tokens each week. It also says the new money will help scale capacity toward 200 megawatts by the end of 2027. Those are not small numbers. But they change the bet. The company Nvidia left behind Groq is no longer mainly asking investors to believe it can beat Nvidia on silicon. It is asking them to believe it can turn inference demand into a durable cloud business while Nvidia sits inside the same story as partner, technology licensee and industry gatekeeper. Business Insider recently described the shift plainly: Nvidia got Groq's technology and talent, and Groq is now integrating Nvidia systems into GroqCloud. That is a different kind of leverage. CoreWeave and Nebius built their names by securing GPU capacity and selling access to developers and enterprises. Groq's route is stranger because it began as a chip company with its own architecture, then watched Nvidia take the most important technical pieces into its own orbit. If you backed Groq for the hardware fight, this is not the same fight. Alex Davis, Groq's chairman and the founder and CEO of Disruptive, framed the company around inference at scale in Groq's June announcement. That attribution checks out, and it is worth reading with both eyes open. Davis is not just an outside commentator describing the market. His firm led the round. That does not make the argument wrong, but it does tell you who is setting the terms of the comeback story. Ross has also been unusually direct about the old Groq. In a July interview with David Senra, he said, "I was a terrible leader," and said his mistakes probably cost Groq three to four years. You don't often hear a founder say that after a $20 billion licensing deal sends him to Nvidia. It gives the whole episode a sharper edge. Groq did not simply lose because Nvidia had more chips, more cash and more customers. It also lost time. The hard part is execution now None of this means Groq is finished. Five million developers is distribution. Thirteen data centers is infrastructure. A funded plan to scale toward 200 megawatts gives customers a reason to keep testing the platform rather than dismissing it as the leftover piece of a bigger Nvidia transaction. Groq Raises Another $350 Million to Build Out Its Nvidia-Powered Neocloud https://startupfortune.com/groq-raises-another-350-million-to-build-out-its-nvidia-powered-neocloud/ Groq raised $350 million on August 17 at a $3.5 billion valuation, the second big check in two months as it completes its pivot from selling AI chips to renting out Nvidia GPU capacity. The move follows Nvidia's $20 billion licensing deal for Groq's LPU technology and the departure of founder Jonathan Ross. - groq raises 350 million funding round valuation https://startupfortune.com/groq-raises-another-350-million-to-build-out-its-nvidia-powered-neocloud/ - groq neocloud nvidia gpu computing rental service https://startupfortune.com/groq-raises-another-350-million-to-build-out-its-nvidia-powered-neocloud/ The hard part is that inference cloud is becoming crowded fast. Nvidia has every incentive to keep the ecosystem close. CoreWeave, Nebius, Lambda and other AI cloud providers are all fighting for the same developers and the same enterprise workloads, and telling the same story about compute scarcity. Groq's advantage has to be more than nostalgia for its LPU pitch. Frankly, calling this just a pivot is too neat. Groq did not walk away from the chip race because a whiteboard said the cloud market looked better. Nvidia licensed the technology, hired Ross and Madra, and turned the remaining company toward a business that depends on capacity and operations, and whether customers still trust it enough to stay. That can work. It is also much less romantic than beating Nvidia at its own game. For a developer choosing where to run inference, the question is practical. Is Groq fast enough and reliable enough, and does the price hold up once the free credits and launch noise fade? For investors, the question is harsher. Can the company left after Nvidia's deal become more valuable than the company that tried to fight Nvidia directly? Groq has fresh money and a real footprint. Now it has to show that the shell Nvidia did not buy is more than a customer funnel for the company it once challenged. Also read: Copper Hits Fresh Record Highs as AI Data Centers Push BHP Past Iron Ore https://startupfortune.com/copper-hits-fresh-record-highs-as-ai-data-centers-push-bhp-past-iron-ore/ • Spotify Labels AI Persona Artists as LinkedIn Buries Its AI Slop Posts https://startupfortune.com/spotify-labels-ai-persona-artists-as-linkedin-buries-its-ai-slop-posts/ • Tesla Readies August Launch of Cybercab, Its Robotaxi Without a Steering Wheel https://startupfortune.com/tesla-readies-august-launch-of-cybercab-its-robotaxi-without-a-steering-wheel/