AI Chipmaker Groq Raises $650 Million in New Funding

AI Chipmaker Groq Raises $650 Million in New Funding AI chipmaker Groq is racing to reinvent itself as a cloud-era AI infrastructure provider, just months after a blockbuster deal with Nvidia stripped away its core leadership and chip IP but handed investors a major payout.

In December 2025, Nvidia struck a non‑exclusive licensing agreement for Groq’s chip technology, in a deal reported to be worth around $20 billion, while hiring away founder and CEO Jonathan Ross, president Sunny Madra, and other key engineers. Nvidia has since integrated the technology into its own lineup, unveiling the Groq 3 LPX inference hardware system at its GTC conference in March.

Left without its founding leadership and exclusive rights to its own language processing unit (LPU) designs, Groq faced an existential question about its future role in the AI ecosystem. The company’s remaining assets were its data centers, software stack, and a nascent cloud business focused on running, rather than training, AI models.

On June 22, 2026, Groq confirmed it had raised $650 million in new funding, roughly six months after the Nvidia deal. The round, led by Disruptive and Infinitum with existing backers reinvesting, is framed as a bet that purpose-built inference chips can still beat general-purpose GPUs as AI shifts toward large-scale model execution.

Groq is now “re-staffing” and leaning into what it calls its “neocloud” inference business, which has grown to 13 data centers across North America, Europe, the Middle East, and Asia-Pacific, serving over five million developers and thousands of AI companies while processing trillions of tokens each week. The company plans to quadruple capacity to 200 megawatts by the end of 2027, deploying its latest hardware and even incorporating Nvidia’s LPX systems into its footprint.

A revamped leadership team is central to the rebuild: Groq has added new executives including a CEO, COO, CTO, and CPO with backgrounds spanning xAI, Meta, and prior cloud software ventures, signaling a shift from chip design toward operating a scaled AI inference cloud.

Continue reading https://foxvector.com/stories/019ef6ce-f86b-0404-732b-3b7b258f4553