Skip to main content

By Groq

Groq

Sub-100ms LPU-based inference for Llama, Mixtral, and other open models.

Filed under Inference Providers. Status: Discovered

On the maker

Groq

Ultra-low-latency inference on custom LPU hardware.

groq.com

What we shipped with it

No notes yet.

We haven’t shipped with Groqyet. When we do, what we learned will land here — terse, dated, honest. The bar is “a real thing we built, including what didn’t work.”

Editorial cadence is bi-weekly. Pieces are mined from real shipping logs, not generated from vendor copy.

Benchmarks

Scores aren’t in yet.

We’re wiring up SWE-bench, Aider Polyglot, and a custom dev-task suite next. Methodology will be public; vendor pre-notification is 48 hours.

SWE-bench Verified
——
Aider Polyglot
——
Custom suite
——