Key facts

Founded
2016
HQ
San Jose, United States
Employees
—
Website
Funding stage
Series A · US$350m
Further facts
  • Licensing: In December 2025 Groq licensed its inference technology to Nvidia on a non-exclusive basis, and founder Jonathan Ross and president Sunny Madra joined Nvidia.
  • Capital: As Groq LLC it raised US$650m of growth capital in June 2026, led by Disruptive and Infinitum, then a US$350m Series A announced on 17 August 2026, led by Disruptive with planned Nvidia participation, at a US$3.5bn valuation.
  • Capital: On 6 November 2026 Bloomberg reported that Groq was seeking new funding that would roughly double its valuation to about US$8bn.
  • Capacity: Groq runs 13 data centres across North America, Europe, the Middle East and Asia Pacific and plans to grow from 54 MW to more than 200 MW in 2027.
  • Developers: GroqCloud serves more than six million developers; on 15 October 2026 it added Meta's Muse Spark open-weight model in limited preview.

Focus areas

LLM inference AI infrastructure Cloud platforms Model APIs AI chips

About

Groq runs GroqCloud, a cloud service for fast, low-cost inference on large language models and other AI models. Jonathan Ross founded the company in 2016 to build its Language Processing Unit chips. After a December 2025 licensing deal with Nvidia, Ross and other staff moved to Nvidia and Groq refocused on running its inference cloud. Adam Winter is chief executive and Matt Eng is chief financial officer.

How they use AI

GroqCloud gives developers API access to hosted models, sold by the token, from data centres in North America, Europe, the Middle East and Asia-Pacific. Groq built its service on its own LPU inference chips, designed for predictable, low-latency execution. The June 2026 funding pays for fitting out its data centres with new inference systems, including Nvidia's LPX platform. A UK data centre with Equinix, opened in early 2026, serves European customers that need lower latency.

Products and services

  • GroqCloud — Inference cloud with API access to hosted AI models
  • LPU inference technology — Groq's inference chip architecture, now also licensed to Nvidia

Key people

  • Adam Winter — CEO
  • Matt Eng — CFO
  • Alan Rice — COO
  • Sinclair Schuller — CTO
  • Rakesh Malhotra — CPO
  • Alex Davis — Chair

Investors

  • Disruptive
  • Infinitum
  • Nvidia (planned)