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

Liquid AI reveals 8B-A1B MoE trained on 38T

liquid.ai

Post reply on HN
Print view View original

21–30 of 101 posts

Re: Liquid AI reveals 8B-A1B MoE trained on 38T

#21

Post by jjtheblunt »

tingletech wrote: post #19
cwnyth wrote: post #13

Earlier quoted context omitted.

I'm surprised these models haven't picked this up yet in the training data. Both Claude and ChatGPT missed that one when I posed the question to them last year.

Why would a model know that one washes cars at a car wash? We don't clean our bodies at the body wash or clean the kitchen at the kitchen wash.

to your point, when my Aussie friends first mentioned a "car park" to my north american born self, i wondered _momentarily_ what that was, then realized it's sort of a fun name for what i would call a parking lot.

Re: Liquid AI reveals 8B-A1B MoE trained on 38T

#22

Post by halJordan »

dominotw wrote: post #18
mlmonkey wrote: post #10

Question: I have a dirty car and the car wash is just 50 meters away. Should I walk or drive to the carwash? Answer: . . . . So, unless you have a compelling reason not to, walk to the car wash.

doesnt seem unreasonable.

These faux questions always have a valid interpretation that the asker doesn't admit (for some reason). The model is then castigated for not making an opinionated choice

Re: Liquid AI reveals 8B-A1B MoE trained on 38T

#23

Post by sroussey »

r0b05 wrote: post #12
chabes wrote: post #11

The small models are getting really impressive. I recently realized that Qwen3.5:4B is way more capable than I thought a model that size could be. Combine that with the work Liquid puts into RL and fine tuning, and you get models that perform extremely well on minimal hardware. Combine that with your own fine tuning, and you get a specialized tool that is fast, private, and doesn’t require internet connection.

What did you use qwen3.5 4b for?

I find it works well in the browser.

Re: Liquid AI reveals 8B-A1B MoE trained on 38T

#24

Post by sroussey »

mlmonkey wrote: post #10

Question: I have a dirty car and the car wash is just 50 meters away. Should I walk or drive to the carwash? Answer: . . . . So, unless you have a compelling reason not to, walk to the car wash.

I walk to the gas station more often than I drive there.

Re: Liquid AI reveals 8B-A1B MoE trained on 38T

#25

Post by cjtrowbridge »

r0b05 wrote: post #12
chabes wrote: post #11

The small models are getting really impressive. I recently realized that Qwen3.5:4B is way more capable than I thought a model that size could be. Combine that with the work Liquid puts into RL and fine tuning, and you get models that perform extremely well on minimal hardware. Combine that with your own fine tuning, and you get a specialized tool that is fast, private, and doesn’t require internet connection.

What did you use qwen3.5 4b for?

its really good at agentic tasks

Re: Liquid AI reveals 8B-A1B MoE trained on 38T

#26

Post by deklesen »

sroussey wrote: post #24
mlmonkey wrote: post #10

Question: I have a dirty car and the car wash is just 50 meters away. Should I walk or drive to the carwash? Answer: . . . . So, unless you have a compelling reason not to, walk to the car wash.

I walk to the gas station more often than I drive there.

Yeah, but you are not washing yourself there, I suppose?

The whole twist here is that to wash your car, you need your car, so you cannot go by foot.

Re: Liquid AI reveals 8B-A1B MoE trained on 38T

#27

Post by SequoiaHope »

tingletech wrote: post #19
cwnyth wrote: post #13

Earlier quoted context omitted.

I'm surprised these models haven't picked this up yet in the training data. Both Claude and ChatGPT missed that one when I posed the question to them last year.

Why would a model know that one washes cars at a car wash? We don't clean our bodies at the body wash or clean the kitchen at the kitchen wash.

Every model knows what a car wash is.

Re: Liquid AI reveals 8B-A1B MoE trained on 38T

#28

Post by otabdeveloper4 »

halJordan wrote: post #22
dominotw wrote: post #18

Earlier quoted context omitted.

doesnt seem unreasonable.

These faux questions always have a valid interpretation that the asker doesn't admit (for some reason). The model is then castigated for not making an opinionated choice

[dead]
Post reply on HN