diffusers

diffusers copied to clipboard

Add single file support for Stable Cascade

3

# What does this PR do? Adds single file support to the StableCascadeUNet, which allows users to load in checkpoints published in the original format. The single file loading logic...

DN6

[`IP-Adapter`] Fix IP-Adapter Support and Refactor Callback for Panorama Pipeline

3

Fixes #7064 for `StableDiffusionPanoramaPipeline`. It looks like most of the work has already been done. `StableDiffusionPanoramaPipeline` class seems to pass the tests of `IPAdapterTesterMixin` in my local setup. Thanks for...

tolgacangoz

[`Tests`] Update deprecated `output_type="numpy"` in test files and fix several typos

1

This pull request updates the deprecated `output_type="numpy"` to `output_type="np"` in the test files, and fixes several typos. @stevhliu

tolgacangoz

Fixes #6126. This pull request includes three commits: - Fix tiling - Add docstring to `decode` method in `ConsistencyDecoderVAE` - Fix typos I tested with this: ```py import torch from...

tolgacangoz

Update Cascade documentation

1

# What does this PR do? There are some issues related to dtypes and torch versions with the Stable Cascade model. This PR: 1. Adds the appropriate information to the...

DN6

Update Stable Cascade Conversion Scripts

3

# What does this PR do? 1. Adds the conversion script for Stable Cascade Lite 2. Updates the conversion script for Stable Cascade to support the `variant` argument Fixes #...

DN6

native_guidance_scale parameter for LCMs in StableDiffusionXLPipeline

16

# What does this PR do? Distilled LCMs don't perform regular class-free guidance. Instead, they pass the `guidance_scale` as conditioning to the U-Net. This is cool because it reduces the...

ivanprado

Fix passing pooled prompt embeds to Cascade Decoder and Combined Pipeline

1

# What does this PR do? We were passing the wrong embedding to the Decoder Pipeline, which resulted in weird artifacts. Fixes # (issue) #7266 ## Before submitting - [...

DN6

[Tracker] use the new model card utilities for saving model cards from the training script

22

With https://github.com/huggingface/diffusers/pull/6729, we shifted to a more resilient and robust way to generate model cards from the DreamBooth training scripts. So, we're keen for the community to help us do...

sayakpaul

ip adapter face id

3

**Is your feature request related to a problem? Please describe.** Can't use ip adapter face id with ip adapter face if I use the community pipeline **Describe the solution you'd...

alexblattner