Thread

Clef: Open-weight decision models, and new RL fine-tuning platform

637 points · 217 comments · jasondavies

mrkn1 · 1790878815 · focus · HN ↗
For smaller scale decision model that runs on CPU, check <a href="https://news.ycombinator.com/item?id=49923223">https://news.ycombinator.com/item?id=49923223
1. fastball · 1790883231 · focus · HN ↗
  tbh saying all these dumb decision models are similar to Jev is like saying markov chains weren't far from GPT-2.
  The value isn't really in the I/O shape, it is in the intelligence combined with the output shape. Every extra ounce of intelligence in these models unlocks additional use-cases. But the converse is also true: a dumb decision model is going to be less useful than using a more intelligent standard LLM.
  That is the appeal of Jev: for certain usage it has more intelligence than some small SOTA LLMs. It is the first decision model that actually feels intelligent (to me).
  1. mrkn1 · 1790891532 · focus · HN ↗
    That makes sense, I agree. I still think there might be use cases when people don't want to use the Jev API, and end on a different trade-off.

Unofficial Hacker News client; not affiliated with Y Combinator.