I’m still not clear if it’s going to deliver the unique layers to you?

If you set a variable layers of 5 for example will it determine what is on each layer, or do I need to prompt that?

And I assume you need enough VRAM because each layer will be effectively a whole image in pixel or latent space… so if I have a 1MP image, and 5 layers I would likely need to be able to fit a 5MP image in VRAM?

Or if this can be multiple steps, where I wouldn’t need all 5 layers in active VRAM, that the assembly is another step at the end after generating on one layer?

▲ jamilton 2 days ago | parent [-]

The linked GitHub readme says it outputs a powerpoint file of the layers.

▲ Llamamoe 2 days ago | parent | next [-]

...of all the possible formats, it outputs.. a powerpoint presentation..? What.

▲

dragonwriter 2 days ago | parent | next [-]

The github repo includes (among other things) a script (relying on python-pptx) to output decomposed layer images into a pptx file “where you can edit and move these layers flexibly.” (I've never user Powerpoint for this, but maybe it is good enough for this and ubiquitous enough that this is sensible?)

▲

djfobbz 2 days ago | parent | prev [-]

Lol, right?!?! I would've expected sequential PNGs followed by SVGs once the model improved.

▲

CamperBob2 2 days ago | parent [-]

That's what the example code at https://old.reddit.com/r/StableDiffusion/comments/1pqnghp/qw... generates. You get 0.png, 1.png ... n.png, where n= the requested number of layers-1.

It'll drop a 600W RTX 6000 to its knees for about a minute, but it does work.

	▲	dvrp 2 days ago \| parent [-]
		I saw some people at a company called Pruna AI got it down to 8 seconds with Cloudflare/Replicate, but I don't know if it was on consumer hardware or an A100/H100/H200, and I don't know if the inference optimization is open-source yet.

▲ oefrha 2 days ago | parent | prev [-]

I don't see the word powerpoint anywhere in https://github.com/QwenLM/Qwen-Image-Layered, I only see a code snippet saving a bunch of PNGs:

  with torch.inference_mode():
      output = pipeline(**inputs)
      output_image = output.images[0]
  
  for i, image in enumerate(output_image):
      image.save(f"{i}.png")

Unless it's a joke that went over my head or you're talking about some other GitHub readme (there's only one GitHub link in TFA), posting an outright lie like this is not cool.

▲

dragonwriter 2 days ago | parent [-]

> I don't see the word powerpoint anywhere in https://github.com/QwenLM/Qwen-Image-Layered,

The word "powerpoint" is not there, however this text is:

“The following scripts will start a Gradio-based web interface where you can decompose an image and export the layers into a pptx file, where you can edit and move these layers flexibly.”

	▲	oefrha 2 days ago \| parent [-]
		Oh okay I missed it, sorry. But that’s just using a separate python-pptx package to export the generated list of images to a .pptx file, not something inherent to the model.