Post

strongly disagree with it being a minor thing. its actually a major advancement in cost and capability for a not insignificant portion of what people are doing with LLMs today. the hype is there for a reason.

antirez@antirez.bsky.social

Jev may have its (narrow) use cases but the hype you see around is the perfect representation of the fact the greatest part of the AI bubble don't know what is important and what is not. Jev is a minor thing happening on AI compared to all the rest, yet the hype exploded.

when you can dramatically reduce both cost and latency of results from models, you can unlock a _lot_ of user-facing outcomes that just wouldn't be possible with current LLMs without ruining user experience

like we are talking about a model that returns in milliseconds results that often beat frontier models with _zero-shot_ prompting. claiming this is "just hype" is imo extremely close-minded

2 reposts85 likes

I think it’s more like “this implementation is nothing special” Seeing that this was a Big Opportunity should probably still get you something tho Maybe Jev the model is not technically impressive or unique and is supplanted entirely in the next few months

That’s part of the pushback yes, also there are a lot of projects that just wouldn’t benefit from it. But there already are other system-1 creatures walking the earth bsky.app/profile/jshe...

Bitey@jsherman999.bsky.social

Laya is a new system 1 model like Jev, but it might be slightly smarter and faster than Jev. Created just moments ago ;) by author of two papers describing the architecture that were published last year.

I just came back to this thread because the release prismml.com/news/bonsai-... of Bonsai 2 happened around the same time. (atleast, it did for my life haha). I like where this is going.....if only I could self hos bonsai 2 on my 3060 (12GB) gpu lol. Thank you for posting this video-super helpful!

prismml.comPrismML — Introducing Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller FootprintTernary Bonsai 2 27B retains 98.2% of Qwen3.8 27B benchmark performance in a 5.9GB footprint, with multimodal and agentic capabilities.