Dhruv Nair comments

Results 214 comments of


                                            Dhruv Nair

Error creating pre-trained FluxPipeline with diffusers "Cannot instantiate this tokenizer from a slow version"

Closing this since it appears to be an environment issue rather than a Diffusers issue.

StableDiffusionPipeline.from_single_file() with CKPT fails: "UnpicklingError: Weights only load failed"

@JemiloII the core issue is similar to what I was talking about here: https://github.com/huggingface/diffusers/issues/9154. There are non-weight serialised objects in the checkpoint file that we don't allow loading via `torch.load`....

[Core] Set `weights_only=True` nicely when we use `torch.load()`.

Hi. Yes, we flipped it because of the security issue. We don't allow setting `weights_only=True` when loading Diffusers models anymore. https://github.com/huggingface/diffusers/blob/76b7d86a9a5c0c2186efa09c4a67b5f5666ac9e3/src/diffusers/models/model_loading_utils.py#L144 Users would have to either convert to safetensors or...

prompt_embeds_scale in FluxPriorReduxPipeline seems to have no effect.

cc: @yiyixuxu