AI Hardware · On-Prem
Groq
Designs its own LPU inference chips; serves Llama + open models at extreme speed.
Foundation proximity: Native (own / self-hosted weights)
Foundational models & how they consume them
PrimaryLlama· self-hosted (LPU) · Native (own / self-hosted weights)
Cloud & datastore
On-PremobjectCompliance
SOC 2Others that build on Meta · Llama
BasetenCerebrasCloudflareDatabricksDellFireworks AIExplore
FAQ
Does Groq use a foundational AI model?
Groq uses Llama (Native (own / self-hosted weights)).
What cloud does Groq run on?
Groq runs on On-Prem.
What database does Groq use?
Groq uses object.
How close does Groq run to the foundational model?
Groq is Native (own / self-hosted weights) — close to the metal, so cheaper tokens and more control over outputs.
Sources
groq.com ↗Also on the map
← Full orbit map (260 companies) · Score your own stack →
Architecture inferred from public sources · confidence high · verify before betting on a detail.