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
Hosting one of these publicly and making the traces public would be fun.
LMAO
i find this is the bare minimum
lowestendgirlfriend
Check sampling settings?
LowEndDistillation
gib settings? I thought you tested it
Works on my machine 🤷
If someone wanna try the model.
https://ai-gurl.click/
Runs in a 16GB KVM on a idle dedi.
Context capped to 4k because reasons.
Can you try running the non XL quant to see if you get better results? Haven't had these sort of issues on my end with XL
It does not work on GreenCloudVPS.
You downloaded the wrong model. You need GGUF quants. I've made these available here:
https://huggingface.co/lowendmicrobacteria/lowendgirlfriend-gguf
lemme try
Should be fixed, the chat templates look broken though.
I said hello and it responded:
So I tried to say hello again and now it's spamming me with emoji, calling me a slut, and trying to make me into its toy. I am alarmed and confused.
Skill (llama.cpp configuration) issue
Really is lowend

It worked on my machine w/ llama.cpp @ Q8/Q4
I don't see the issue?
share settings
as soon as my GPUs get back online i’m distilling this into a chinese model
This one starts.
We uploaded Python script.
It returned thousands of tildes
~~~~.Hello yoursunny.
It sounds like you need to check your sampler settings.
Thank you-
This is true love