ES version is available. Content is displayed in original English for accuracy.
Advertisement
Advertisement
⚡ Community Insights
Discussion Sentiment
77% Positive
Analyzed from 1371 words in the discussion.
Trending Topics
#jev#llm#llms#typesafe#models#more#data#same#laya#open
Discussion (32 Comments)Read Original on HackerNews
It’s a bit faster and bit cheaper, but this is compared to LLM. The consistency was nice to see, BUT, as someone who trained NLP models prior to LLMs, it’s just BERT with more data. I can see why people would want ready made one shot classifier, and I can see the value of sending multiple classifier in one call, but I wouldn’t call it breakthrough. And I believe many labs will replicate it in no time and might have it as part of their harness.
I see it as a wake up call for the tech community to go back to basics for most tasks instead of relying solely on generic LLMs.
My company specializes in statistical long document text classification, but nowadays we mainly work with audit trail requirements because we got tired of hearing complaints about our 5 example learning curve. Seems like the industry standard is telling an llm to label and telling an llm to eval, and crossing your fingers that it’s correct.
It’s hard to justify several months to business when there is something off-shelf ready to use and doesn’t require domain specialists to run.
- the addition and standardization (with incomplete coverage) of the solution of adding typing to Python
- how much people are re-discovering the value of performance + typing (e.g. Rust)
then I'm going to take a small leap and extrapolate that the trend will be similar here.
The equivalent of the "one off script in python" will be the LLM, and the long term stable and maintainable solution will be something much more structured and focused like Jev.
We ripped out custom homegrown ml models that were developed in last 10 yrs and put an llm in its place. Its the opposite of wasteful. Even local gemma models are vastly superior.
> I see it as a wake up call for the tech community to go back to basics for most tasks instead of relying solely on generic LLMs.
I always say the cheapest LLM request is no request at all.
[1] https://arxiv.org/abs/2507.18546
[2] https://arxiv.org/abs/2311.08526
[3] https://github.com/urchade/GLiNER
That's a pretty big limitation, I would argue, unless I'm misunderstanding and it can be worked around easily somehow? I'm surprised it isn't surfaced more prominently in the comparison.
This Jev waitlist that Typesafe AI are utilising is surely going to raise questions pretty soon - it's hard to sell this to bosses when it looks like a pop-up restaurant
[1] https://arxiv.org/abs/2503.23303
Whereas the other guy went through the unglorious but formerly respectable path of publishing software and papers for other professionals to look at. A year ago.
We're in a bad place where the latter looks less reliable than the former.
(EDIT: I'm not saying the research here is in fact the same as what "Jev" is doing; and Jev is in fact more "product shaped." But I think it's important to temper the hype and back up and focus on the fact that this whole industry is built on research by both academics and enthusiasts ... first ... and gold rushes can often bulldoze over those people who are focused primarily on making-doing-researching instead of fundraising-hyping-promoting. That's not good.)
Laya seems to be focused on sales/conversations?
Reading quickly about TypeSafe, it seems to be about creating _type-safe_ outputs from AI tools for downstream systems to consume, we actually have a system in production that's probably a glove-fit for that, it's for scanning receipts to be ingested into a system and we also have other systems in a sales-pipe that isn't too far off Laya but still sounds more pertient to TypeSafe.
You did a special case well, but just because they cover (perhaps badly) that case doesn't mean that it's the same thing.
“Claude, roast this noob, tell him that his model isn’t novel or frontier —”
both in unison “— and make no mistakes!”
It’s all so tiresome
The implosion of hype after the .com crash was actually kind of a ... relief.
Now Laya promises another speed up and it's open source. Tbh if it can't run on a CPU I anyway want to buy it from an inference provider. Managing gpus in production is a non trivial problem.
What I also wondered about Jev is how different it is from something like tabular foundation models. They seem to overlap in use cases. Which then leads to the question, what is actually learned? A lot of people in machine learning spend time to making things explainable and always struggled to move beyond data induced biases.
Having it open source is awesome as fine tuning might give additional performance on the task we care about.