Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

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 görüntüleme • 5 gün önce •via X (Twitter)

32 Yorum

Jaro profil fotoğrafı
Jaro5 gün önce

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

David Valerio Gilmore profil fotoğrafı
David Valerio Gilmore5 gün önce

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

Dominik profil fotoğrafı
Dominik5 gün önce

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

g023 profil fotoğrafı
g0235 gün önce

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

Dragos Roua profil fotoğrafı
Dragos Roua5 gün önce

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

Jeramy profil fotoğrafı
Jeramy5 gün önce

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

srujan profil fotoğrafı
srujan5 gün önce

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

n0geegee profil fotoğrafı
n0geegee5 gün önce

tried it, trashed it

Hyre profil fotoğrafı
Hyre5 gün önce

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

johnny smitthy profil fotoğrafı
johnny smitthy5 gün önce

What's the model?

M M profil fotoğrafı
M M5 gün önce

No local model matches 4o in anything.

The Singularity Layer profil fotoğrafı
The Singularity Layer5 gün önce

Let's add this on Singularity Network.

Andrew Rose profil fotoğrafı
Andrew Rose5 gün önce

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?

𝖄𝖊𝖓 profil fotoğrafı
𝖄𝖊𝖓5 gün önce

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

A Concerned Human profil fotoğrafı
A Concerned Human5 gün önce

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

Alade Yessoufou profil fotoğrafı
Alade Yessoufou5 gün önce

Which model is that ?

Juan Carlos Vez Vázquez 👣 Artesano en Mk y Ventas profil fotoğrafı
Juan Carlos Vez Vázquez 👣 Artesano en Mk y Ventas5 gün önce

Awesome!

Akash Sardar profil fotoğrafı
Akash Sardar5 gün önce

Lfm 3 my guess i have tried it too

Donovan Baney profil fotoğrafı
Donovan Baney5 gün önce

It is certainly one of the greatest models.

Adi R 🇺🇲🇮🇱🇺🇦💙🌎👽 profil fotoğrafı
Adi R 🇺🇲🇮🇱🇺🇦💙🌎👽5 gün önce

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!

Velsa 🌹💙 profil fotoğrafı
Velsa 🌹💙5 gün önce

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

Nick profil fotoğrafı
Nick5 gün önce

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

ALUCAB 🇪🇺 profil fotoğrafı
ALUCAB 🇪🇺5 gün önce

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

Taylor W profil fotoğrafı
Taylor W5 gün önce

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

Jonny D profil fotoğrafı
Jonny D5 gün önce

how do you recommend using a model like this?

Aercrist profil fotoğrafı
Aercrist4 gün önce

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.

Fin Boots profil fotoğrafı
Fin Boots5 gün önce

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

Ash profil fotoğrafı
Ash5 gün önce

Liquid cooked with this one

Sven Meyer profil fotoğrafı
Sven Meyer5 gün önce

on what inference engine do you run it ?

E1ev3n. profil fotoğrafı
E1ev3n.5 gün önce

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.

落落 profil fotoğrafı
落落5 gün önce

Can it reliably run locally on a Pi?

Piotr Pabis profil fotoğrafı
Piotr Pabis5 gün önce

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

Benzer Videolar