QWEN IMAGE EDIT 2511 can do (N)SFW by itself

I didnt know that 2511 could do that without waiting for the AIO model.

https://redd.it/1pvco6n
@rStableDiffusion

A small teaser for the upcoming release of VNCCS Next!
https://redd.it/1pvi5w5
@rStableDiffusion

How to stack 2 or more LoRA's (Like a style and character) and get good results with Z-Image

https://redd.it/1pvjntk
@rStableDiffusion

Qwen-Image-Edit-2511 workflow that actually works
https://redd.it/1pvj4u6
@rStableDiffusion

Why there are soo few Z-image character Lora's?

For me, Z-image have proved to be the most efficient checkpoint (in every sense) for 8gb vram. In my opinion, it put others checkpoints to shame in that category.

But I can't find characters Lora's for it. I understand it is fairly new, but Flux had Lora's exploding in the early days.

There is a reason for that?

https://redd.it/1pvmlij
@rStableDiffusion

Just curious, but can we use Qwen3-VL-8B-Thinking-FP8 instead of 2.5 version in the new Qwen Image Edit 2511?



https://redd.it/1pvv63l
@rStableDiffusion

TurboDiffusion: 100–200× Acceleration for Video Diffusion Models
https://github.com/thu-ml/TurboDiffusion

https://redd.it/1pvx6cu
@rStableDiffusion