Groq is a company that runs AI models on its own chip, the LPU, and sells fast inference through the GroqCloud API.
In plain words
Many AI models run on GPUs, chips first built for graphics. Groq designed a different processor, the Language Processing Unit, specifically for the maths of running AI models, and keeps memory on the chip. Developers reach it through an API that mostly follows OpenAI's format.
A quick example
Imagine a voice assistant where every pause feels awkward. The developer points their existing OpenAI client at Groq's address with a Groq key, so replies stream back quickly enough to feel like conversation.
Where you'll meet it
- Apps where response speed matters, such as voice and agents.
- Running open models through an OpenAI-style API.
Related
Sources
- DocsGroq LPU AI inference technology, Groq · read 29 Sept 2026
- DocsOpenAI compatibility, Groq · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 29 Sept 2026. Spotted an error? Report it.