Загрузка видео...

Не удалось загрузить видео

На главную

This is a 1.7B param model btw, using the browser, making html files, and managing itself in Pi. This model hits 4o levels on most benchmarks lol Great for workflow automations.

121,160 просмотров • 5 дней назад •via X (Twitter)

Комментарии: 32

Фото профиля Jaro
Jaro5 дней назад

Yeah this lfm model is amazing - works on my android with 3gb ram at 20t/s

Фото профиля David Valerio Gilmore
David Valerio Gilmore5 дней назад

nice - I hadn't tried the new LFMs with more agentic capabilities. going to look into training hybrid edge/cloud semantic routing for it in our next ARC checkpoint

Фото профиля Dominik
Dominik5 дней назад

4o levels? So great at causing psychosis but not much else?

Фото профиля g023
g0235 дней назад

pretty epic. Also the 2B with vision punches way above its weight

Фото профиля Dragos Roua
Dragos Roua5 дней назад

You can even use it on a Raspberry Pi: It’s the stock one, without webGPU kernels.

Фото профиля Jeramy
Jeramy5 дней назад

I’ve been experimenting with this one too. For laptops without great chips, it’s pretty amazing what it can do!

Фото профиля srujan
srujan5 дней назад

heard a bunch of terrible stuff about lfm3, but glad to see someone doing some good stuff with this will def look into lfm3 automation into a new agent focused browser im working on

Фото профиля n0geegee
n0geegee5 дней назад

tried it, trashed it

Фото профиля Hyre
Hyre5 дней назад

a 1.7B model writing its own html files would’ve sounded made up last year

Фото профиля johnny smitthy
johnny smitthy5 дней назад

What's the model?

Фото профиля M M
M M5 дней назад

No local model matches 4o in anything.

Фото профиля The Singularity Layer
The Singularity Layer5 дней назад

Let's add this on Singularity Network.

Фото профиля Andrew Rose
Andrew Rose5 дней назад

This is the size I care about for office automations. My 14B on a 3080 runs around 18 tok/s with some layers on CPU, a 1.7B that can drive a browser leaves the whole card for context. How's it hold up on longer multi-step stuff?

Фото профиля 𝖄𝖊𝖓
𝖄𝖊𝖓5 дней назад

Yup, my poor man’s local agent 🥹 works like a charm👌

Фото профиля A Concerned Human
A Concerned Human5 дней назад

It's pretty good but it's quality of work could use work

Фото профиля Alade Yessoufou
Alade Yessoufou5 дней назад

Which model is that ?

Фото профиля Juan Carlos Vez Vázquez 👣 Artesano en Mk y Ventas
Juan Carlos Vez Vázquez 👣 Artesano en Mk y Ventas5 дней назад

Awesome!

Фото профиля Akash Sardar
Akash Sardar5 дней назад

Lfm 3 my guess i have tried it too

Фото профиля Donovan Baney
Donovan Baney5 дней назад

It is certainly one of the greatest models.

Фото профиля Adi R 🇺🇲🇮🇱🇺🇦💙🌎👽
Adi R 🇺🇲🇮🇱🇺🇦💙🌎👽5 дней назад

It's Mac browser I assume? My iPhone browser refuses to run even 325m param model - site. Runs great on Mac though. Feedback is most welcome!

Фото профиля Velsa 🌹💙
Velsa 🌹💙5 дней назад

fiquei curioso, essa interface bem feita, é de qual app? é nativa?

Фото профиля Nick
Nick5 дней назад

Gonna go check it out. AI for exploration is awesome.

Фото профиля ALUCAB 🇪🇺
ALUCAB 🇪🇺5 дней назад

this is great. any writeup or link i can read ? I am trying to run it in collaboration with picoclaw for example but it loses context also for simple action requests and conversations

Фото профиля Taylor W
Taylor W5 дней назад

Demo videos need bass-heavy instrumental dance soundtracks for me to take them seriously.

Фото профиля Jonny D
Jonny D5 дней назад

how do you recommend using a model like this?

Фото профиля Aercrist
Aercrist4 дней назад

How do you build custom silicon for Liquid AI’s LFM3-1.7B? The tri-hybrid architecture (Conv + Sliding Window + Global Attention) requires a completely different ASIC design than pure transformers. Here’s how to architect an edge AI chip for it.

Фото профиля Fin Boots
Fin Boots5 дней назад

Pretty insane this is even possible. How’s tool calling capability?

Фото профиля Ash
Ash5 дней назад

Liquid cooked with this one

Фото профиля Sven Meyer
Sven Meyer5 дней назад

on what inference engine do you run it ?

Фото профиля E1ev3n.
E1ev3n.5 дней назад

The browser part is what got me. I run a small model on my own machine for the boring jobs, and the moment it stopped needing the cloud my whole workflow got calmer. A 1.7B model that makes its own HTML files and manages itself is the first time the small end of the scale has felt like the useful end to me.

Фото профиля 落落
落落5 дней назад

Can it reliably run locally on a Pi?

Фото профиля Piotr Pabis
Piotr Pabis5 дней назад

And it’s amazing at vision, I was completely stunned

Похожие видео