‹ BackHN Continuity

Thread

Jeff – Jev-compatible 0.8B decision models, trained at home, ~30 ms

575 points · 225 comments · firelex

  1. adrithmetiqa · · focus · HN ↗
    Forgive my lack of understanding but how long before Jev type functionality is just built straight into all frontier models?
    1. tbeseda · · focus · HN ↗
      If I had to guess, it's already built and is just waiting on Product's/Marketing's desk. How do you position this without looking like your roadmap is being determined by newcomers? Probably don't want to adopt the same verbiage+acronyms - but also can't be seen to be just sherlocking features.
      1. seizethecheese · · focus · HN ↗
        I think Apple has demonstrated that shipping second has essentially no negative impact if your product is seen as higher quality.
        1. bigyabai · · focus · HN ↗
          I think Nvidia has demonstrated that shipping first is a multi-trillion dollar opportunity if you don't shy away from a challenge.
          1. throwaway27448 · · focus · HN ↗
            Chip manufacturing intrinsically comes with one hell of a moat. There's not much parallel in software.
            1. slashdev · · focus · HN ↗
              That’s kind of funny because Nvidia’s biggest moat is arguably CUDA, the software ecosystem around their chips
              1. mcmcmc · · focus · HN ↗
                CUDA is a lock-in moat, the infrastructure needed for chip manufacturing is a barrier-to-entry moat. Two different things.
                1. AtlasBarfed · · focus · HN ↗
                  If cuda is an API, and llms make apis effortless, then how big of a moat is cuda?
                  1. latentsea · · focus · HN ↗
                    It's becoming less of one. Previously I would have shied away from buying an AMD card because of CUDA, but with local LLMs getting good enough to be usable and frontier models becoming as good as they have, I bit the bullet and got an R9700 for local inference. Dealing with working around CUDA used to be more of a manual process, but when you can point an agent at it and get stuff working, it's dramatically less painful and scary than it used to be. Plus, at least in ComfyUI and local LLMs I'm finding support for AMD has gotten really good. Lately I've been witnessing a lot of people using agents to write custom kernels for RDNA4 and improving performance dramatically.
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.