Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

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 Aufrufe • vor 5 Tagen •via X (Twitter)

32 Kommentare

Profilbild von Jaro
Jarovor 5 Tagen

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

Profilbild von David Valerio Gilmore
David Valerio Gilmorevor 5 Tagen

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

Profilbild von Dominik
Dominikvor 5 Tagen

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

Profilbild von g023
g023vor 5 Tagen

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

Profilbild von Dragos Roua
Dragos Rouavor 5 Tagen

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

Profilbild von Jeramy
Jeramyvor 5 Tagen

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

Profilbild von srujan
srujanvor 5 Tagen

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

Profilbild von n0geegee
n0geegeevor 5 Tagen

tried it, trashed it

Profilbild von Hyre
Hyrevor 5 Tagen

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

Profilbild von johnny smitthy
johnny smitthyvor 5 Tagen

What's the model?

Profilbild von M M
M Mvor 5 Tagen

No local model matches 4o in anything.

Profilbild von The Singularity Layer
The Singularity Layervor 5 Tagen

Let's add this on Singularity Network.

Profilbild von Andrew Rose
Andrew Rosevor 5 Tagen

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?

Profilbild von 𝖄𝖊𝖓
𝖄𝖊𝖓vor 5 Tagen

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

Profilbild von A Concerned Human
A Concerned Humanvor 5 Tagen

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

Profilbild von Alade Yessoufou
Alade Yessoufouvor 5 Tagen

Which model is that ?

Profilbild von Juan Carlos Vez Vázquez 👣 Artesano en Mk y Ventas
Juan Carlos Vez Vázquez 👣 Artesano en Mk y Ventasvor 5 Tagen

Awesome!

Profilbild von Akash Sardar
Akash Sardarvor 5 Tagen

Lfm 3 my guess i have tried it too

Profilbild von Donovan Baney
Donovan Baneyvor 5 Tagen

It is certainly one of the greatest models.

Profilbild von Adi R 🇺🇲🇮🇱🇺🇦💙🌎👽
Adi R 🇺🇲🇮🇱🇺🇦💙🌎👽vor 5 Tagen

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!

Profilbild von Velsa 🌹💙
Velsa 🌹💙vor 5 Tagen

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

Profilbild von Nick
Nickvor 5 Tagen

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

Profilbild von ALUCAB 🇪🇺
ALUCAB 🇪🇺vor 5 Tagen

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

Profilbild von Taylor W
Taylor Wvor 5 Tagen

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

Profilbild von Jonny D
Jonny Dvor 5 Tagen

how do you recommend using a model like this?

Profilbild von Aercrist
Aercristvor 4 Tagen

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.

Profilbild von Fin Boots
Fin Bootsvor 5 Tagen

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

Profilbild von Ash
Ashvor 5 Tagen

Liquid cooked with this one

Profilbild von Sven Meyer
Sven Meyervor 5 Tagen

on what inference engine do you run it ?

Profilbild von E1ev3n.
E1ev3n.vor 5 Tagen

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.

Profilbild von 落落
落落vor 5 Tagen

Can it reliably run locally on a Pi?

Profilbild von Piotr Pabis
Piotr Pabisvor 5 Tagen

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

Ähnliche Videos