I invented the TPU at Google. The Tensor Processing Unit. Then I left because I knew I could build something faster. That chip is the LPU.
Speed is the feature. When you can run inference 10x faster than anyone else, every AI application changes. Real-time conversations. Instant code generation. Things that feel sluggish on GPUs feel instant on Groq.
NVIDIA builds GPUs for everything. We built a chip for one thing: running language models as fast as physically possible. Specialization always beats generalization.
Our demo went viral because people could see the difference. They typed a question and the answer appeared faster than they could read it. You don't need a benchmark to feel that.
The name is Groq with a Q. Not Grok with a K. We were here first. Our chip was in development before Elon named his chatbot. Just to be clear.