I don't know why, now i can't use RVC Model in this version, i knew that my GPU Vram is 2GB Vram. but, before this version, i can use any RVC Model. It can be fix, or should i download previous version?
#Can't use RVC model in MMVCServerSIO v.1.5.3.17b
honest tide
honest tide
I don't know why, now i can't use RVC Model in this version, i knew that my GPU ...
in the picture say
return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None, non_blocking)
torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 20.00 MiB (GPU 0; 2.00 GiB total capacity; 152.20 MiB already allocated; 0 bytes free; 154.00 MiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
spec :
Intel i5-1135G7 2.40GHz (8CPUs)
Nvidia Geforce MX330 2GB Vram
cosmic gobletBOT
Ayo? @honest tide level 1 !!! 
golden radish
in the picture say
return t.to(device, dtype if t.is_floating_point() or t.is_c...
it literally says 2 GB is too low, 4 GB is the possible minimum but 6-8+ GB is more recommended
honest tide
it literally says 2 GB is too low, 4 GB is the possible minimum but 6-8+ GB is m...
i knew. but, in previous version, i can use the model without any problem
should i downgrade the version?
or others solution?
golden radish
i knew. but, in previous version, i can use the model without any problem
it's supposed to be the opposite, the 1-year old Nvidia driver would throw OOM error but the newer one would only cause severe performance drop
honest tide
1 year ago, i use this voice changer in original windows 11 and i got bad performance
when i change to atlas os, it can run without any performance drop
but now, i only can use beatrice model (like japanese male/female voice)
golden radish
when i change to atlas os, it can run without any performance drop
then stick on that linux distro 