My understanding was that the weights were the "essence" of the training.
I think it's a bit misleading to present them as "executables".
I agree that open weights is a bit of marketing mumbo jumbo but I wouldn't say they are akin to a closed source binary.
That being said I was just reading an article about LLM sleeper agents and their trigger words... So you can hide stuff on training and it's fairly hard to spot with just the weights.
But again it's not really like a black box executable. And I've seen many great models that successfully builds on top of an open weights model.
In the end I much prefer open weights to the very popular "'''openAI""" that have opaque training and weights...
I think Bazzite is the "easiest". But I think it would be very difficult to tinker for someone not used to Linux. It's the plug and play option.
For me the fact that bazzite tries to be immutable is a very good plus for stability on the long run. And somehow fits well for gaming on Linux. The drawback is that these immutable distro are hard to tinker with if you dont have experience with immutable package managers and so on.
CachyOS has maybe a more traditional structure but should offer good performance too.
There is also Nobara and Pop OS.
I'm on PoPOS but it's too recent for me to give feedback for gaming. But it should work well too.
Sometimes people require medical treatments at home.
My grandma had for quite some time the visit of a nurse at home to replace a catheter. They would use one of those medical waste bins right at home with the patient in these case.
Or maybe this is just a veterinary?
Edit : probably not a vet, I dont think vets use tongue depressors.
I think for someone that is very knowledgeable In a project they would probably somehow now if there is vibe coding. I think this will affect brand new projects but not that much of the older codebase. Even think it might enable finding old bugs in old open source codebase.
That's probably my main issue with Windows : Its ability to change settings on its own.
I feel like I have almost not control over my OS. It's not a tool that helps me do stuff, it's a dumb assistant that thinks he understands what I'm trying to achieve.
"Oh you plugged a PS5 Dual Sense controller I see, let me switch your microphone to the controller even though you are actively already using another one".
"Oh you put your computer in sleep before going to bed? Let me switch it on In the middle of the night to update, we will call that a mandatory maintenance because you can't disable that feature".
I really need to spend more time on my Linux boot rather than this shitty W10 setup".
My understanding was that the weights were the "essence" of the training.
I think it's a bit misleading to present them as "executables".
I agree that open weights is a bit of marketing mumbo jumbo but I wouldn't say they are akin to a closed source binary.
That being said I was just reading an article about LLM sleeper agents and their trigger words... So you can hide stuff on training and it's fairly hard to spot with just the weights.
But again it's not really like a black box executable. And I've seen many great models that successfully builds on top of an open weights model.
In the end I much prefer open weights to the very popular "'''openAI""" that have opaque training and weights...