[Go to site: main page, start]
Start building

GroqPlatform

Build your perfect stack. GroqMetal provides infrastructure, GroqCore adds inference, and GroqAssured adds enterprise controls. Each includes the layers below.

Start Building
  • Infrastructure

    GroqMetal

    Dedicated bare-metal infrastructure, tuned for speed and reliability, with full control in your hands.

  • Inference

    GroqCore

    A tested inference stack that turns dedicated capacity into production-ready performance, no infrastructure expertise required.

  • Control

    GroqAssured

    Enterprise-grade governance, auditability, and control layered on top, so scale never comes at the cost of trust.

  • 256 LPUs per rack
  • 40 PB/s SRAM bandwidth
  • 1,000 tokens/sec/user
  • 128 GB of on-chip SRAM per rack
  • 315 PFLOPS of FP8 inference compute

Groq operates fast, reliable inference at massive scale with fine-grained control.

When released, each NVIDIA Groq 3 LPX rack connects 256 next-generation LPU accelerators to NVIDIA’s Vera Rubin to deliver low-latency, large-context inference for production AI.

Globally online

Now operating 13 data centers across four continents

  • US-1 — Liberty Lake, USA
  • US-2 — St Paul, USA
  • US-3 — Dallas, USA
  • US-4 — Houston, USA
  • CA-1 — Kamloops, Canada
  • CA-2 — Vaudreuil-Dorion, Canada
  • CA-3 — Calgary, Canada
  • EU-1 — Vantaa, Finland
  • AU-1 — Sydney, Australia
  • EU-2 — London, UK
  • SA-1 — Riyadh, Saudi Arabia

Put inference
to work

Start Building