JoshuaChris

wouldn't that be just inefficient and it would be redundant to make a mount profile or each file to have a custom/forced compression when btrfs-defrag could change the attr and...

bump, as it would be interesting.

would this work on older nv gpus? for instance one that throws errors for whisper torch. also does the gpu need to have the entire model in memory? (sorry I'm...

sorry, I meant to ask if it is possible to split a larger model. as looking at the standard gpt-j ggml usage goes from 16 to 32GB of ram, I...