Groq

by Groq · paid · Last verified 2026-03-17

Groq is a semiconductor and AI inference company that has built the Language Processing Unit (LPU), a custom chip architected specifically for sequential token generation. Groq's inference API consistently delivers the fastest publicly available LLM inference speeds—often exceeding 800 tokens per second for large models—at competitive pricing, making it the preferred choice for latency-sensitive agentic applications.