Thanks! Isn’t it better to train the embeddng with the model I expect to use it with?
Thanks! Isn’t it better to train the embeddng with the model I expect to use it with?
I think Automatic1111 runs on most CPUs, but you need a lot of ram.
Cars are for unrefined masses. Men of culture prefer trains and piblic trasportation.
I think the real gamechanger was Vulkan. OpenGL was just not suited for this.
Back in 2010s, I had to open Windows to game. Now all games work out of the box with Linux, thanks to wine, vulkan and dxvk.
There is a pretty big FuckCars community on Lemmy.
WallStreetBets
Automatic1111 is amazing, isn’t it?
I want to try it out before making a major investment.
Awesome!
Auto1111 might be trying to load multiple models at the same time, which it does not have room for.
SDXL is very memory hungry. Most base models are around 6-7 GB, which doesn’t leave much room for anything else.
Not suprising. Linux is usually faster, that is whe backend of every internet service uses Linux.
Pipewire is amazing. Linux had issues with Bluetooth audio that Pipewire finally fixed.
Yo ho yo ho 🏴☠️🏴☠️🏴☠️🏴☠️
It clearly shows a major update in signups
Valve almost makes me believe in capitalism.
It’s in development. I think there is beta functionalty already that you can try.
Is that why India has the best programmers?