Build your own decision model

Join the discussion

• Hacker News
• To me the Jev moment isn't about Jev / Typesafe. It's articles like these. It's the letting a million Jevs bloom.

  The bubble will come from a realization that a lot of people can train these models. AI Researchers will become more diffuse, work at more companies, and building a Jev or fine-tuning an LLM isn't some trillion dollar frontier lab exercise, but increasingly just something developers do.

  by softwaredoug
• Saw the question "Where would you most likely find a bat?", it occurs to me there's an innate tension, do we want the llm to be factually correct or do we want it to be more average human like? As an average human being not a sme on the subject my first instinct answer would be cave as well. I think it's reasonable to expect trainning on the aggregate of the internet means it would arrive at the same answer.

  Edit: The context of the question does indeed make it sound more like the animal bat. The other answers sound more like gotchas to me.

  by utbabya
• Does someone have examples of interesting stuff that has been built utilizing Jev/decision models? The way this is hyped up surely there must be some good stuff?
  by sva_
• This is very cool. If you are looking for something similar but more lightweight, that you can run (and train) on CPU, try out Jeffy: https://jeffyclassify.com/

  On GitHub: https://github.com/nicobrenner/jeffy

  by nico
• So "decision model" is the new cool term for classifiers? You know, that thing that already existed decades ago
  by mastazi
• This is a great way to start but the perf of LLM models like Qwen are not ideal for local execution. I have adapted Laya (pure decision model) to run in a browser and I am able to get responses under 200ms. Give it a try: https://wexare-ai.github.io/browser-laya/
  by sachaa
• I saw Jev and without understanding it said to myself, “I can build that!” And brainstormed some weird Alex trebek jeopardy generator I called trebek, a bun ran typescript app that you give it an input, it decides if it’s a category, question or answer then generates what’s missing. Trained a SQLite-vec database on the English language for a few days with a small qwen embedding model to add vec embeds to the db then ripped the cord on the embeddings before I started feeding my llm the proper specs for Jev and now i have a cool jeopardy generator that now doubles as a Jev clone classifier with a custom v1 endpoint for system 0 or whatever it is. Level understanding here, fun experiment though ^>^ and produces useful outputs
  by sn0n
• As an MLE who has been failing to get anyone interested in classifiers for many years, the hype around Jev makes me scream internally.

  Yes, I get that a zero-shot classifier is more convenient than the traditional kind, it's very cool. Kind of. But then again plain LLMs have been perfectly cheap and serviceable as zero-shot classifiers for quite some time now, so again I'm back to my internal screaming.

  by howunfortunate