NEWS: We are the FIRST among AI start-ups and incumbent providers to run #LLM Llama-2 70B at 100 tokens per second (T/s) per user, using Groq LPU™ systems! Read more at http://
groq.link/100tps and if you're interested in a private demo, reach out to us at contact@groq.com.

→ View original post on X — @groqinc