hlky

@souvikg544 the PR is finished already, just waiting on reviews.

Documenting a workaround for peak memory we found for visibility: ```python tensors = {} with safe_open(filename, framework="pt") as f: keys = list(f.keys()) for key in keys: with safe_open(filename, framework="pt") as...

@Nerogar On Windows 

Basic version of #10112 Detail-Daemon, original has more [advanced calculation for adjustment](https://github.com/Jonseed/ComfyUI-Detail-Daemon/blob/90e703d3d3f979438471c646a5d030840a2caac3/detail_daemon_node.py#L283-L289) ```python import torch from diffusers import StableDiffusionXLModularPipeline from diffusers.pipelines.stable_diffusion_xl.pipeline_stable_diffusion_xl_modular import ( StableDiffusionXLTextEncoderStep, StableDiffusionXLDecodeLatentsStep, StableDiffusionXLInputStep, StableDiffusionXLAutoSetTimestepsStep, StableDiffusionXLAutoPrepareLatentsStep, StableDiffusionXLAutoPrepareAdditionalConditioningStep, )...

See https://github.com/huggingface/diffusers/issues/10112#issuecomment-2519752589 it needs to pass a different sigma to model than is used for sampling

@yiyixuxu Yes I'll work on that

I'd have to dig into FlowMatch a bit more in the community implementations, but it looks like there's little difference between `EulerDiscrete` and `FlowMatchEulerDiscrete` so `FlowMatchEulerDiscrete` could be integrated into...

The design surrounding samplers/schedulers in Diffusers has always been a pain point, spanning all the way back to 2022 where the separation of `model` and `scheduler` complicated the inclusion of...

@yiyixuxu Option 3 sounds good to me, I've made those changes here [a00f104](https://github.com/huggingface/diffusers/pull/12792/commits/a00f1048e0a214fd09764cff5dd3afb0300fba1d). Let me know if you have any further comments, I will add support for `from_single_file` next.

`from_single_file`: ```python import torch from diffusers import ZImageControlNetModel from huggingface_hub import hf_hub_download controlnet = ZImageControlNetModel.from_single_file( hf_hub_download( "alibaba-pai/Z-Image-Turbo-Fun-Controlnet-Union", filename="Z-Image-Turbo-Fun-Controlnet-Union.safetensors", ), torch_dtype=torch.bfloat16, ) ```