Does an open-weight decision model beat a hosted one? Jev vs. Laya

Does an open-weight decision model beat a hosted one? Jev vs. Laya astgl.com

Laya vs Jev for typed AI decisions: why I run the open-weight model on my Mac, how it beat my router 37/40 to 33/40, and what that does not prove.

15 points•Jmeg8r•8 days ago•5 comments•

#jev #beat #laya #weight #open weight #hosted #weight decision #decision model

5 comments

orsorna8 days ago

For anyone who was piqued by the title but couldn't stomach the unedited LLM blogpost, you can read the similarity AI-genned docs for Laya directly: https://github.com/NandhaKishorM/laya

ozozozd8 days ago

Man, it shows that someone was getting paid by the word.

Thank you!

CROON_tv8 days ago

Similar result in production for me. I compared Jev with a general LLM (DeepSeek) as the decider for when to cut live speech into sentences, on the same 226 segments. With Jev the longest held segment was 17 s. With DeepSeek it was 40 s.

mkrishnan7 days ago

I tried both with a synthetic dataset (1000 items), Jev scored 98% vs Laya 15%.

f30e3dfed1c97 days ago

The flood of posts related to this thing does not seem organic.

Read the full thread on Hacker News →

5 comments

Related stories