Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Introducing System One Models and Jev

typesafe.ai

Post reply on HN
Print view View original

451–460 of 513 posts

Re: Introducing System One Models and Jev

#451

Post by anentropic »

mortsnort wrote: post #336

I am confused why they say it is not an LLM and then in the documentation it is shown as being an LLM derivative. The documentation makes it sound like they're taking a pretrained LLM and then giving it their unique post-training. How is that not an LLM? FAQ: Is Jev just a smaller LLM? Jev is neither small nor an LLM, hence being off the intelligence Pareto curve. Image in documentation: https://mintcdn.com/ts-docs/a…

Yeah, it must be an LLM for some definitions of LLM

It seems to take two forms of context input: 'state' and 'questions'

https://docs.typesafe.ai/concepts/state

> State can be as simple as a string

> State can also be a JSON object or array containing related context, examples, and other information that helps the model answer the associated questions.

> The state contains the content and supporting facts.

The state seems to be schemaless, while the questions determine the output schema.

Re: Introducing System One Models and Jev

#455

Post by sebzim4500 »

wg0 wrote: post #329
copperx wrote: post #199

Earlier quoted context omitted.

I'm sure you've thought of self-driving. How does the model work in that space?

But the real problem in self driving isn't the decision making but object description. That is, computer vision if with cameras. Decision making isn't that of a bottleneck I suppose.

99.9% of the time you are right, but in the long tails (construction works, major road damage, temporary road markings that contradict permanent signs) I could see this kind of architecture being useful.

Re: Introducing System One Models and Jev

#456

Post by smusamashah »

rana3g wrote: post #366
nullbio wrote: post #354

Earlier quoted context omitted.

More like: I need to ...? -> Open-weight model. I'm sure someones working on this as we speak using an open-weight LLM base (Qwen or something would be a perfect fit). This sort of task is a perfect fit for a very small model capable of semantic parsing. You can get away with a LOT less parameters without all the autoregressive generation and long-context reasoning.

you don't say - https://huggingface.co/harshatheg/Qwen-2.5-1B-RLCD

https://x.com/harshagundal/status/2100044305536889015 tweet by the author

Re: Introducing System One Models and Jev

#457

Post by LeBit »

wg0 wrote: post #319
cfowles wrote: post #268

Wasn't really till seeing this home assistant demo they have ( https://www.loom.com/share/18c4dbcf8db546dfb2d7f2ef018e78e4 ) that the value really clicked for me. Seems really cool.

That's good. Side note - just like most people don't need an intelligent personal assistant to manage and respond their emails and book their flights, most people also don't need smart homes. Century old toggle switches are more than enough in a 3 room apartment or 5 room house unless you have a mention.

I once made the mistake a buying a light bulb that needed a phone app to turn on and off.

That was my light bulb moment.

Re: Introducing System One Models and Jev

#458

Post by shifto »

qingcharles wrote: post #321
wg0 wrote: post #319

Earlier quoted context omitted.

That's good. Side note - just like most people don't need an intelligent personal assistant to manage and respond their emails and book their flights, most people also don't need smart homes. Century old toggle switches are more than enough in a 3 room apartment or 5 room house unless you have a mention.

My primary beef with smart home (having tried it) is that every person that visits your home ends up confused about some element of it. A light switch that goes up and down is universally understood.

I have a hugely integrated smart home. 2 rules; 1. Everything can still be manually operated in the way people expect it to. 2. Local only.

Re: Introducing System One Models and Jev

#459

Post by edot »

hangrymoon01 wrote: post #386
edot wrote: post #235

Very cool! Can you explain when I would use this vs. training a standard ML model on my data? Suppose I had a fraud dataset with features like customer ID, amount, merchant, online or in-person, etc. - I can't imagine that a general model like Jev would predict this more accurately or cheaply than even a basic XGBoost model trained on my dataset (one that I could build in a few minutes by asking Codex to build it). W…

you will need to collect data for every decision/usecase and then train a model. But this can be used for different use cases with just a prompt. Founders response to a similar question on X: https://x.com/CompleteSkeptic/status/2100067328620896408?s=2... pasting it here: zero-shot + general == programmable I would assume any extreme scale narrow task could then be fine-tuned for, but we'll see - I suspect putting it…

Thanks. I do concede it’s very general but that is a double-edged sword. I don’t need a general fraud identification algorithm. I need an accurate one. If I have another classification task I’ll train another model for that task.

Re: Introducing System One Models and Jev

#460

Post by stiiv »

jacobgold wrote: post #81

First, congrats to the team on launching something genuinely interesting and new. Seems like a more accurate title would be "Jev: Trading general purpose generation for fast typed inference" or something like that. This is interesting, but the speed comparison seems misleading? A generative model that can output code in a Turing-complete language can do anything a computer can do. Jev can only generate structured out…

> but it's nothing like the code generating models we're all using today for code and automation.

Is this true? Code is structured output. At the very least it seems like a question of degree rather than kind.

While the LLMs we're using today are limited to sequenced text, it seems that a model like Jev could excel at coding on a more structural level (factoring, controls) by working within the constraints of an actual language specification and supplemental domain model. I don't know, though -- maybe that's too deep and complex.

Post reply on HN