the model is getting loaded in ram instead of vram
Thanks for asking the question. Please refer here for loading the model to GPU and data parallel.
· Sign up or log in to comment
