‹ BackHN Continuity

Thread

Kev: Tiny Jev-like family of decision models built on top of Qwen3.5

462 points · 211 comments · tosh

  1. nico · · focus · HN ↗
    If you only need classification, and you can provide some training data, you can ask Codex/Claude to build an embeddings + logistic classifier model for you

    For emails, I get 95% accuracy with this method, with only 50-100 examples for training

    Training the model takes less than 5 minutes on a CPU

    The resulting model is <1MB, and inference is sub 100ms

    Some other cool things about this approach:

    * the model doesn’t train on some “ideal” or general classification, instead it learns your preferences

    * the model runs on pretty much any mobile device and can be retrained online on the device

    * privacy, the whole training and inference is 100% local, no data goes anywhere (except whatever you feed codex/claude while building the model)

    Note: to do a more general test, I made a classifier for the Banking77 dataset. The model is <10MB, trains in <30s on CPU and gets 94.5% accuracy, which puts it in the top 5?models by accuracy for that set (the best one is at 94.86%, but it’s 350MB in size and takes hours to train on a GPU).

    1. 0x457 · · focus · HN ↗
      I built a whole thing that collects data, trains classifiers, exports models and dataset just for that. Claude writes me a terraform file that contains shape of the classifier and dataset. For images it can create datasets based of another dataset (crop this region from images that have these labels).

      Originally it was so I can label data to fine-tune a VLM, but now a few tiny classifiers that run in milliseconds on cpu.

      Now its collecting data to make a domain specific BERT and do what Jev does.

      1. swyx · · focus · HN ↗
        > Now its collecting data to make a domain specific BERT and do what Jev does.

        this misses the point of jev somewhat - the point is that this is a foundational, general purpose classifier model - see some good sources <a href="https:&#x2F;&#x2F;x.com&#x2F;mparakhin&#x2F;status&#x2F;2101683565520199887?s=12" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;mparakhin&#x2F;status&#x2F;2101683565520199887?s=12

        1. 0x457 · · focus · HN ↗
          Missing the point that they marketing, but so far I saw it perform good at some tasks, and pretty bad on others. I don&#x27;t think it&#x27;s all that general purpose they claim it to be.
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.