News for hackers. Stuff that matters.
Re:Show HN: A working Nintendo in ~1GB of Qwen weights by vkgcr on Saturday October 31, 2026 @10:41AM (#50316642)
But wouldn't that be horribly inefficient? Instead of executing code by transforming it into a set of weights and running inference on them, wouldn't it be easier to just execute code as... code?

There are no comments on this comment yet.