RNGD Server: A Game-Changer for Enterprise Customers
The RNGD Server has officially hit the market, offering enterprise customers the opportunity to delve into the world of large language models (LLMs) like LG’s EXAONE LLM. This cutting-edge technology is making waves across various sectors such as electronics, finance, and telecommunications.
LG AI Research conducted rigorous testing on the RNGD Server and the results were impressive. The server boasted a 2.25 times better performance per watt for LLMs compared to a GPU-based alternative. The compute density of a RNGD-powered rack also stood out, enabling it to generate a staggering 3.75 times more tokens for EXAONE models within the same power constraints as a GPU rack.
The Rise of AI Chip Startups: FuriosaAI and Beyond
While the AI chip startup scene is bustling with activity, FuriosaAI has managed to capture attention with its innovative solutions. In a bold move, the company turned down a lucrative $800 million acquisition offer from Meta, the parent company of Facebook. Despite this, there are other notable players in the field specializing in AI inference, such as Cerebras, Graphcore, Groq, and SambaNova, among others.
According to Addison Snell, CEO of Intersect360 Research, the market for AI accelerators is fiercely competitive. While these companies may not pose a significant threat to industry giant Nvidia just yet, the dynamic nature of the market leaves room for them to secure notable victories in the near future.
