tildebyte

I know that this will probably sound a certain way, but: is this even English? Personally, I'm sick of Torch's horrible technical writing... `torch.cuda.set_per_process_memory_fraction` > Set memory fraction for a...

AFAICT, `77` is baked into T5, or at least *HuggingFace's* T5. `77` is hard-coded in '[t5.py](https://github.com/deep-floyd/IF/blob/2ad4dff7cde3cce91f237270a3cc81cae4578015/deepfloyd_if/modules/t5.py#L92)' Bumping that number up (by manually editing 't5.py'; I tried '144') simply results in...

@rentainhe; Your text-to-image prompts are waaaayyy too long (see e.g. #108) - otherwise, nice guide! 👍

This: https://github.com/deep-floyd/IF/discussions/89 Also the following | aspect: | 2:3 | prompt: | 3/4 portrait, resin bust, Mad Joker model, clear irises, 3d, pixar style, ultra perfect composition, liquid, detailing fluid,...

@phalexo; I'm running all the full models in full resolution on a 48G VRAM RTX A6000 instance on RunPod[1]. What are you using? [1] This is not meant as some...

@phalexo; > not sure why two busts come out `aspect='2:3'` is super important If you give it a wide aspect (3:2) the model will fill the space with *something* -...

This isn't available when using 🤗Diffusers pipelines; you have to run as indicated in [Run The Code Locally](https://github.com/deep-floyd/IF#run-the-code-locally). Notice the args to `dream()` in the "I. Dream" section - you...

#66 claims that it's possible to run inference with IF using only 6G VRAM, but I have not tested it myself

This has nothing to do with Stable Diffusion

You fundamentally misunderstand what "open source" means. https://opensource.org/osd Also, keep in mind that *this repo* may be open source, but there is no reason on Earth why the *trained model*...