Running LLM models may use 75% less RAM, according to researchers who developed a new way to run LLM models. This will make AI cheaper to run on your own computer.
Every week, something amazing happens in AI.
What used to be a once-in-a-decade event is now just a few days away at any given moment.
This time: 75% memory reduction in LLMs. Inching ever closer to local machines.https://t.co/rkP4yOWkRk