Groq is seeking an $8 billion pre-money valuation in a new Series B funding round, Bloomberg reported on November 6, citing investor documents. The target is about two and a half times the $3.15 billion pre-money valuation at which Groq closed a $350 million Series A led by Disruptive in August. Both figures exclude the new money raised.
Executive chairman Alex Davis has approached existing backers with the round. In an investor letter Davis argued that Groq can increase its margins without leverage, a pitch aimed directly at the week’s dominant anxiety: debt funded data center operators such as CoreWeave sold off sharply as investors questioned borrowing heavy AI infrastructure models. Davis projected that Groq’s revenue could rise eightfold in 2027 and reach $2 billion of EBITDA by 2028. The new money is intended to meet demand for inference compute under short and medium term contracts.
The valuation path is the story. Groq was valued at $6.9 billion in September 2025. In December 2025 Nvidia struck a roughly $20 billion non-exclusive licensing deal for Groq’s inference technology and hired founder Jonathan Ross, president Sunny Madra and much of the senior engineering team. The company that remained rebuilt itself as an AI cloud operator, was certified as an Nvidia Cloud Partner, and repriced in August at roughly half its September 2025 mark. It has since secured $1.5 billion from Saudi Arabia for regional infrastructure and launched data center operations in Helsinki.
Nothing about the round is settled. No raise amount and no lead investor have been named, Groq and Disruptive declined to comment, and the $8 billion figure is a target in a pitch, not a priced deal. Whether investors pay it, in a week when the Nasdaq fell 5.5 percent in a single session on AI debt fears, is the cleanest available test of how much of the AI infrastructure repricing is about leverage rather than about demand.