siraxe

vibe debugging says issue is this [SDTrainer.py](https://github.com/user-attachments/files/24195975/SDTrainer.py) duplicates embedding to use when caching

> yes seems to work quite nice. My rusty 3090 might be too weak for the BF16, so my test was with the fp8. Will give the bf16 a try...

@myprivacygithub use branch https://github.com/kijai/ComfyUI-WanVideoWrapper/tree/longcat

Yeah it might have been sage1 , I had: ``` pytorch 2.8.0+cu129 sage 1.0.6 Python 3.13.6 ``` Did a clean comfy install + updated nvidia drivers (had 576 > installed...

Yeah 16 steps feels much better , so slow tho But still pretty amazing that you can gen this locally now with not much effort. https://github.com/user-attachments/assets/f2987b27-2844-4d11-9cec-029118591275 https://github.com/user-attachments/assets/0affc618-b50b-4b3a-a088-41c799017dff

Looking at what it does is just passing motion frames from one part to another , so if subject it out of frame it might not return in same shape...

Ah yes , more Instagram slop will be created :D Some stuff is hard to control still , like camera pause or damage/explosion shape https://github.com/user-attachments/assets/308334ae-21ce-480e-a345-c6679d22253d

same with qwen , I think the way loras are applied is broken , no OOM without applying loras - no loras = 21gb vram - any node that applies...