Lincoln Stein

When a glyph's left end is positioned on one of the fine scrolling virtual screens, then the label is not initially visible unless the user scrolls off to the left....

LoRA patching optimization

3

## Summary This implements a scheme in which the original weights are not restored after LoRA patching, saving a second or so per LoRA. This only occurs under the following...

Cannot free VRAM after loading a quantized model

1

### System Info ```Shell - `Accelerate` version: 0.31.0 - Platform: Linux-5.15.0-79-generic-x86_64-with-glibc2.35 - `accelerate` bash location: /home/lstein/test_ckpts/SD3/.venv/bin/accelerate - Python version: 3.10.12 - Numpy version: 2.0.0 - PyTorch version (GPU?): 2.3.1+cu121 (True)...

[MM] Make controlnet image processors use MM cache system

4

## Summary This PR changes the loading of most controlnet image processors to use the model manager caching system. This should result in a performance improvement when multiple different control...

Add node to read watermarks

2

## Summary This adds an invocation to read the invisible watermark created by the AddWatermark invocation. ## Related Issues / Discussions ## QA Instructions  1. Create...

[bug]: Disable model Convert button when an IP-adapter is selected

1

### Is there an existing issue for this problem? - [X] I have searched the existing issues ### Operating system Linux ### GPU vendor Nvidia (CUDA) ### GPU model RTX...

[MM] add API routes for getting & setting MM cache sizes

1

## Summary This PR adds the following model manager-related API routes: * `get_cache_size` Fetch the maximum size of the RAM or VRAM cache. * `set_cache_size` Set the maximum size of...

## Summary This implements alpha-level support for generating images with Stable Diffusion 3. To use: 1. Install the Diffusers version of SD3, located at `stabilityai/stable-diffusion-3-medium-diffusers`. For convenience, this repo has...

### Is there an existing issue for this? - [X] I have searched the existing issues ### Contact Details _No response_ ### What should this feature add? @RyanJDick encountered a...

## Summary We're starting to see fine-tuned CLIPEmbedding models for improved FLUX performance appearing in the wild. For example: https://huggingface.co/zer0int/CLIP-GmP-ViT-L-14 . However, these fine-tunes are formatted as single "checkpoint"-style files...