RuntimeError: CUDA out of memory

[ldyeax]

RuntimeError: CUDA out of memory. Tried to allocate 58.00 MiB (GPU 0; 11.00 GiB total capacity; 10.04 GiB already allocated; 0 bytes free; 10.21 GiB 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

I've tried closing everything else that uses any GPU memory, but it always says "0 bytes free".

Trying to use edit_app.py in PowerShell on Windows 10 conda 22.11.1 2080Ti 11GB

[holynski]

Hey, we recently updated the README to clarify that the default configuration needs >18GB of GPU memory, so your device probably doesn't have enough memory.

There are, however, a few other ways of running our model with less memory, listed at the bottom of the page here.