News for hackers. Stuff that matters.
Re:Show HN: Run 500B+ Parameter LLMs Locally on a Mac Mini by lazerlapin on Saturday March 14, 2026 @11:43AM (#47377806)
With 1.58-bit ternary quantization, you may think you're running a big model but really you're just running a "mini" version of it

There are no comments on this comment yet.