another-ai

In the gradio_app/app.py load one model at a time

1

In the gradio_app/app.py: 1- load one model at a time, 2- c_dtype = torch.bfloat16 3- miss a generator=generator on def generate_prior(...) as an example see: https://github.com/another-ai/stable_cascade_easy With a nvidia rtx...

Slow on gpu with 12 gb vram, but there's a solution...

3

The speed increases considerably if you load one model at a time. It can also work with video cards with 12 gb vram (even at 1280x1536 resolution and similar) So...