Groq
Groq is an AI / GPU Cloud company founded in 2016 and based in Mountain View, United States.
Founded in 2016 and headquartered in Mountain View, Groq develops custom AI inference silicon and cloud services. The company's core product is the Language Processing Unit (LPU), a proprietary ASIC designed for large language model inference with deterministic, sub-millisecond latency and high token throughput. Groq monetizes through GroqCloud, a managed inference platform using a pay-per-token model that serves over 2.5 million developers, and GroqRack, an on-premises solution for enterprise deployments. The company operates as a private entity with 201–500 employees.
Groq's LPU architecture addresses a fundamental constraint in AI deployment—inference latency and throughput consistency. By moving from GPU-based inference to custom silicon optimized for token generation, Groq enables production AI applications requiring deterministic performance at scale. The breadth of adoption across GroqCloud's developer base and enterprise demand for on-premises solutions establishes the company as a specialized competitor in the AI inference infrastructure sector.
groq.com ↗·AI / GPU Cloud·
United States·private 
