AI inference specialist Groq has closed a $350 million Series A funding round at a $3.5 billion post-money valuation.
Groq is no longer positioning itself primarily as an AI chip challenger to NVIDIA. Following a major technology licensing agreement and a leadership transition, the company is rebuilding around a different opportunity: operating AI infrastructure at scale. Groq has announced a $350 million Series A round led by Disruptive, with planned participation from NVIDIA, valuing the company at $3.5 billion. NVIDIA’s commitment had not been finalized at the time of the announcement, and the round remains subject to customary closing conditions. Combined with the $650 million raised in June, Groq has now secured $1 billion in recent funding.
Founded in 2016 by Jonathan Ross, who previously helped launch Google’s TPU program, Groq originally developed the Language Processing Unit (LPU)—a specialized processor designed for fast and predictable AI inference. This strategy initially placed the company in direct competition with NVIDIA. The trajectory shifted after Groq signed a non-exclusive technology licensing agreement with NVIDIA in December 2025. While remaining independent, Groq pivoted from developing proprietary AI chips to building cloud infrastructure, a transition previously covered by Wowtale alongside its $650 million June raise.
Today, Groq operates 13 data centers across North America, Europe, the Middle East, and Asia-Pacific, serving more than six million developers, Fortune 500 enterprises, and thousands of AI-native companies. The new capital will fund medium- and large-scale NVIDIA accelerated computing clusters for AI training and inference. The company plans to expand its total capacity from 54 megawatts to over 200 megawatts by 2027.
The current $3.5 billion valuation represents approximately a 49% decline from the $6.9 billion valuation assigned in September 2025. Direct comparisons between the two rounds carry limitations, as the earlier figure reflected Groq’s role as a proprietary inference chip alternative to NVIDIA. Investors are now pricing in the company’s ability to operate inference infrastructure, scale data center capacity, and convert its extensive developer base into paying cloud customers. Consequently, labeling this round a “Series A” is unconventional given Groq’s history of large financings; in this context, the designation signals a reorganized business model rather than traditional early-stage venture financing.
Groq is entering a rapidly capitalizing neocloud market where rivals are raising billions to secure GPUs, power, and data center footprint. Australian infrastructure provider Firmus recently closed a $2 billion round at a $10.5 billion valuation, while U.K.-based Nscale raised $2 billion in Series C funding at a $14.6 billion valuation. Publicly traded CoreWeave received a $2 billion investment from NVIDIA and aims to deploy more than 5 gigawatts of AI factory capacity by 2030. By both valuation and planned capacity, Groq remains smaller than these competitors. Its competitive edge rests on its experience running specialized inference systems and its network of over six million developers. Long-term success will ultimately hinge on hardware utilization rates, power economics, software stack quality, and the execution of multi-year customer contracts.