@noroute@yoasif local LLM execution times can be very fast on recent consumer hardware. No need to send anywhere, just like their translation - do it all on-device.
As an example, with no optimization or GPU support, my @frameworkcomputer (AMD) generates around 5 characters/sec from a 4 gigabyte pre-quantized model.
@shellsharks yeah, not a fan of the way snap sucks disk space. But went with it for laptops because of Steam support. Still Fedora on server so have feet in both camps too