正在加载视频...

视频加载失败

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 次观看 • 5 天前 •via X (Twitter)

32 条评论

Jaro 的头像
Jaro5 天前

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

David Valerio Gilmore 的头像
David Valerio Gilmore5 天前

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 的头像
Dominik5 天前

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

g023 的头像
g0235 天前

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

Dragos Roua 的头像
Dragos Roua5 天前

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

Jeramy 的头像
Jeramy5 天前

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

srujan 的头像
srujan5 天前

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 的头像
n0geegee5 天前

tried it, trashed it

Hyre 的头像
Hyre5 天前

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

johnny smitthy 的头像
johnny smitthy5 天前

What's the model?

M M 的头像
M M5 天前

No local model matches 4o in anything.

The Singularity Layer 的头像
The Singularity Layer5 天前

Let's add this on Singularity Network.

Andrew Rose 的头像
Andrew Rose5 天前

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 的头像
𝖄𝖊𝖓5 天前

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

A Concerned Human 的头像
A Concerned Human5 天前

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

Alade Yessoufou 的头像
Alade Yessoufou5 天前

Which model is that ?

Juan Carlos Vez Vázquez 👣 Artesano en Mk y Ventas 的头像
Juan Carlos Vez Vázquez 👣 Artesano en Mk y Ventas5 天前

Awesome!

Akash Sardar 的头像
Akash Sardar5 天前

Lfm 3 my guess i have tried it too

Donovan Baney 的头像
Donovan Baney5 天前

It is certainly one of the greatest models.

Adi R 🇺🇲🇮🇱🇺🇦💙🌎👽 的头像
Adi R 🇺🇲🇮🇱🇺🇦💙🌎👽5 天前

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 🌹💙 的头像
Velsa 🌹💙5 天前

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

Nick 的头像
Nick5 天前

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

ALUCAB 🇪🇺 的头像
ALUCAB 🇪🇺5 天前

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 的头像
Taylor W5 天前

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

Jonny D 的头像
Jonny D5 天前

how do you recommend using a model like this?

Aercrist 的头像
Aercrist4 天前

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 的头像
Fin Boots5 天前

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

Ash 的头像
Ash5 天前

Liquid cooked with this one

Sven Meyer 的头像
Sven Meyer5 天前

on what inference engine do you run it ?

E1ev3n. 的头像
E1ev3n.5 天前

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 的头像
落落5 天前

Can it reliably run locally on a Pi?

Piotr Pabis 的头像
Piotr Pabis5 天前

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

相关视频