So quick question: What is this?
I mean, I can guess that it’s a weird new Facebook thing. But unless it’s also a new Windows thing hardwired into the OS like the Edge/Cortana bullshit, I don’t really care what FB is up to. :/
At least meta AI actually releases most of their models, and publishes research papers about the ones they don’t.
“Who are those horrible orange creatures over there?”
“Why, those are the Grunka Lunkas. They work here in the Slurm factory.”
“Tell them to fuck off!”
Where are you guys running into this? I thought for the most part tech people dropped anything associated with meta
your defacto group messaging choice
Don’t throw all of us under that bus, please and thanks
Since it’s what everyone else uses, it’s what I need to use. Good luck getting everyone to switch to another app
Some people are ‘forced’ to use WhatsApp due to it being the messaging platform of choice for a lot of people.
I put it in quote marks. Obviously I’m not actually forced to use it, but my entire family uses it and it’s the dominant messaging platform in Indonesia - to the point where people get annoyed with you if you send regular SMS or just end up replying to you on WhatsApp. A lot of businesses large and small also exclusively use it, so you’re causing yourself a lot of annoyance if you actively avoid WhatsApp. It’s a crap situation, but it’s reality in some countries.
The quantized model you can run locally works decently and they can’t read any of it, which is nice.
I use that one specifically https://huggingface.co/lmstudio-community/Meta-Llama-3-8B-Instruct-GGUF/blob/main/Meta-Llama-3-8B-Instruct-Q4_K_M.gguf
If you’re looking for a relatively user-friendly software to use it, you can look at GPT4All (open source) or LM Studio.
If you’re ready to tinker a bit i can recommend Ollama for the backend and Open web UI for the frontend. They can also both run on the same machine.
The advantage is that you can use your GPU to compute, which is a lot faster.
