Inference Deployment

[hajkeoadf]

Hi author! Thank you for your great work.

I am curious about the need for GPU resources to deploy inference of MagicBrush. Is it possible to run on 4060 with 8GB VRAM?

[drogozhang]

Hi,
I don't know the details but I roughly remember the GPU memory cost is small. So I think it's very likely to run it on a GPU with 8GB VRAM. Happy to know whether it works later from you.