New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

Comments
Yep, GAs can break the score instantly. (Is it?) Good thing - there were not many GAs this time, so less chance to break the scores
I am pretty sure that it will change soon. Most of the OG hype squad is busy this year. And, it's a good thing I guess, because they prolly have much bette rpaying jobs now. I also think xpress couldn't be active this time because last time he was WFH
I have already told everyone, I don't scrape anything in home lab, home lab is only a proxy for the Stats, everything is handled on the VM. But since the VM has more than 1GB RAM (4GB to be exact), maybe some model would fit?
But it would be very slow though 
Oh, I had it all wrong then. Anyways, I think maybe a 1b model might work, maybe. Worth a try though, given it just doesn't end up using 100% all the time.
I once tried to run a T2I generator on a 1GB VPS. My expectations were quickly put to rest.
good morning. trying to do some work while also looking in here.. 👀👀👀
I am totally not into AI stuff, do you know any AI Text detection tools I could look into?
oh well.. a picture is worth like 40 words at least on average.. maybe you should program that in.. or at least make sure that pics are different and such..
Most of us right now:-

I have heard good things about the lightweight qen models on Ollama. Has been quite some time having checked it out, but, let me see if I can quickly dig up soemthing.
i saw but decided to let my silence speak for me.. "no" is more of an explanatory answer
Makes sense, since no one commented on the thread about I posted a month ago too
Nope, a picture is not your own written content
And this was introduced when we had a lot of palettes in the thread. These have blown up the comment count, but it was great to see the similarity numbers telling the truth 
can just use something API online to query once and get the score.. unless those cost money
I guess LET is all about "being human." Infact, now that I say it, I see very little discussion on AI-related topics here.
I do feel like last/previous years, there was at least one giveaway a day and it alternated between US and "EU" timings. I'm willing to bet it will pickup next week when there's even more folks than the usuals in this thread looking.
there is no rush really..
Yup, but, it would definitely be a fun project to see if a small AI model can run on an LE VPS.
Invoice #18901877
Hello, I would like to double the bandwidth.Order placed on November 12, 2025
My Order Number is: #18901877
"Thanks!
🚀 DOUBLE YOUR BANDWIDTH ON ALL 11.11 DEALS - just comment below your order # within this thread, and we'll double up your bandwidth allocation!"
Good morning folks! How was everyone's day? Did I miss anything?
@MrEd Haven't looked into tools yet, but, Ollama has llama3.2:1b which might just work on your VPS - https://ollama.com/library/llama3.2
With Open-WebUI or even with Ollama's CLI, I believe you can setup API and use it to analyse text. Worth a try.
What's up?
No actually!
i mean, i was just being modest..
https://en.wikipedia.org/wiki/A_picture_is_worth_a_thousand_words
Maybe, Dustin is also warming up, just like us. But, to be fair, last year, there were atleast 5-6 people who were aiming vigorously for something on the prize list.
Hi, I'm back!
Not yet, Pudlez. May have just proven that someone else has taken your curse up this year. I think it's me BTW.
nothing happened in a while here, right? at least 24hrs
Ollama is pretty easy to use
Hey, dude. Back home?
We need to make much more noise here.