-
Hi,
i am wondering if there is any known problems with using stable-fast on few-steps generation models such as [SDXL Lightning](https://huggingface.co/ByteDance/SDXL-Lightning)? My experience is …
-
How about DMD2 on img2img? Other methods like Hyper-SD, SDXL-Lightning don't perform well on img2img tasks, resulting in blurry images?
-
### Bug description
It seems not good to use BitsandbytesPrecision directly as shown at front pages of Lightning.
what should I do to quantize SDXL and make it saved after quantization?
Codes a…
-
Is it possible to implement support for Turbo SDXL models or Lightning SDXL or TensorRT?
-
### Is there an existing issue for this?
- [X] I have searched the existing issues and checked the recent builds/commits of both this extension and the webui
### What happened?
PuLID Vram leak whil…
-
Hi.
I use **realisticStockPhoto_v20** on Fooocus with **sdxl_film_photography_style** lora and I really like the results.
Fooocus and other gradio implementations come with settings inputs that I …
-
```
ValueError: Incompatible shapes for attention inputs:
query.shape: torch.Size([8, 1500, 10, 64])
key.shape : torch.Size([2, 1500, 10, 64])
value.shape: torch.Size([2, 15…
-
Hi Oleg, and thanks for the great portal.
My 2080 is not working at expected speeds.
I have 9-10it/s, but Comfy only returned a picture every 5 seconds.
I tied to add --hihram in comfy argumen…
-
Is there a plan to support faceid
-
### What happened?
Caching of a large dataset (5,000 images) is causing the RAM to increase at 0.1GB increments every 1 or 2 seconds.
This was with SDXL.
I have just created a conda environment wit…