
About Groq
Founded in 2017, Groq offers AI chips with ultra-low latency and high throughput. Groq LPU can run 70-billion-parameter enterprise-scale language models at more than 100 tokens per second; estimated speed advantage of 10x-100x compared to other systems.