‹ BackHN Continuity

Thread

Turning GLM-5.3-Flash into a Jev-like decision model

138 points · 59 comments · flxflx

  1. ricardobeat · · focus · HN ↗
    Everyone is doing this to emulate Jev, but...

    I took a random book excerpt with 23,000 words (±30k input tokens) and used it as context. Jev still responds in 800ms, sometimes 500ms. That's in the neighbourhood of 20-50,000 tok/s prefill, which is obviously not possible with normal LLMs, not even Cerebras is this fast.

    1. ericpauley · · focus · HN ↗
      This has “/dev/null as a service” vibes…
      1. walrus01 · · focus · HN ↗
        magic 8 ball as a service.
        1. nacs · · focus · HN ↗
          Is the Magic 8 ball immune to hallucination as Jev is?
          1. IshKebab · · focus · HN ↗
            Yes. It's exactly as immune to hallucination - it can only produce answers from a small fixed set.
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.