Every story and comment from the Hacker News API, in a layout borrowed from 2001.
A comment on Refact LLM: New 1.6B code model reaches 32% HumanEval and is SOTA for the size. Parent comment
But its way beyond the point where its going to help LLMs. CPU RAM is already "too slow" in machines big enough for multiple NVMe SSDs.
There are no comments on this comment yet.
But its way beyond the point where its going to help LLMs. CPU RAM is already "too slow" in machines big enough for multiple NVMe SSDs.