Jev claims to be frontier intelligence. Laya, while claiming to be "the open source version of Jev", is using a tiny open weight model with a tiny context window. Anyone who has experimented with tiny models knows that they are far from "frontier intelligence". It's not plausible that Laya could be "the open source version of Jev", with "frontier intelligence", when it is using these tiny models.

Also, the paper that OP is referring, is not describing anything that sounds like a generalist classifier (which is what Jev is). Their paper describes a tailored solution to one specific business problem. I'm sure it has some similarities with Jev, but it's still a completely different thing, and I'm confused why OP is claiming it to be the same thing.

If you don't believe me, just open the PDF and read the abstract.

But is there any proof that Jev is frontier intelligence?

It thinks there are two Rs in strawberry.

It fails to assign probabilities to die roll outcomes or coin flips.

When used as an LLM it thinks it is Qwen.

Apparently reordering the list of possible answers can change assigned probabilities by up to 20%.

I’ve seen claims that it struggles to play tic tac toe.

So far there is a lot of evidence that it behaves exactly like a tiny open weight model. The only argument against this claim is “trust me bro” claims from it’s author.