Groq Joins NVIDIA's Cloud Partner Program to Dominate AI Inference
Groq earns NVIDIA Cloud Partner certification, validating its ability to design and operate NVIDIA accelerated computing infrastructure to the highest standard globally.
- Groq joins NVIDIA Cloud Partner (NCP) program, earning certification to design and operate NVIDIA accelerated computing to NVIDIA's own reference architecture standards.
- Certification formalizes a deepening relationship that began with a non-exclusive inference technology licensing deal in December 2025 and NVIDIA's participation in Groq's $650M Series A.
- Groq operates 13 data centers globally across North America, Europe, the Middle East, and APAC, serving 6M+ developers generating trillions of tokens weekly.
- NCP status clears the path for Groq to deploy the latest NVIDIA accelerated computing in its existing data centers, targeting 200 MW of capacity by end of 2027.
- Existing GroqCloud customers gain access to more capacity and newer inference technology without any code changes.
- Groq's LPU architecture — which keeps model weights on-chip to eliminate HBM memory bottlenecks — remains the core technical differentiator behind the company's speed claims. Read the announcement.
Groq has joined the NVIDIA Cloud Partner (NCP) program, earning certification that it can design, deploy, and operate NVIDIA accelerated computing infrastructure to NVIDIA's own reference architecture and operational standards. For an inference-focused neocloud that has spent years arguing speed and reliability are its differentiators, the certification is external validation of claims its customers have been making for a while.
What the NCP badge actually means
The NVIDIA Cloud Partner program certifies GPU-as-a-service providers as compliant with standards NVIDIA dictates, offering a minimum guarantee of architectural compliance and operational maturity on a cloud platform. The requirements cover the full stack of AI cloud infrastructure services and operations for partners delivering GPU capacity on NVIDIA's behalf.
The NCP reference architecture is a blueprint for building high-performance, scalable, and secure data centers capable of handling generative AI and large language models. It lets partners cut the time and cost of deploying AI systems while ensuring compatibility and interoperability across hardware and software components. Earning the certification means Groq has proven it can run NVIDIA infrastructure the way NVIDIA itself would run it.