Groq is an AI chip startup founded by former Google employees. The company has introduced an AI acceleration chip named LPU, which significantly speeds up the inference and generation of large models through technological innovation, achieving speeds up to 10 times that of GPUs. This is primarily due to the adoption of high-speed SRAM storage technology and an architecture design that minimizes memory access. Users can run various large models such as Llama and Mixtral on the LPU. The introduction of LPU helps to further optimize the performance of large models and may be used to enhance the response speeds of applications like voice assistants and AI writing.
Large model generation speed becomes a bottleneck, a team of former Google employees develops a new chip LPU

新硅
This article is from AIbase Daily
Welcome to the [AI Daily] column! This is your daily guide to exploring the world of artificial intelligence. Every day, we present you with hot topics in the AI field, focusing on developers, helping you understand technical trends, and learning about innovative AI product applications.