v3: Cleaner, more eye-catching, more vibrant... otherwise it has the same features as v2. all settings are the same...
v2: Created using Qwen Image 2512. Produces extremely realistic and detailed images. Provides cleaner outputs compared to the first version. Ideal for NSFW content but not very creative for hardcore scenes.
v1: This is my first qwen share. It was created to achieve realistic and cinematic style visuals. It may not be suitable for anime-style visuals. It is EXTREMELY suitable for NSFW style content. Despite its large size, it is extremely fast. I created it for fast use. If you are going to use it with LORAs, keep the LORA weight values very low. Otherwise, the images you create may be distorted.
The most suitable usage I have found:
Steps: 6-8
cfg: 1
Resolution: 1328x1328 or 1440x1440 resolution and 2:3, 3:4, 16:9, 1:1, and all other ratios
Sampler: Euler a + beta/ bong tangent/ DDIM Uniform
Sigma Shift: 2-3
Using these settings, it takes about 25-30 seconds to generate an image at 1440x1440 resolution and 8 steps on my RTX3090 computer.
IMPORTANT: If you use settings other than the ones I provided (different samplers other than Euler-A, clip skip:2, high-value LORAs, high sigma shift values, etc.), you will most likely get corrupted images.
Description
FAQ
Comments (12)
Do you have a workflow? I'm not able to get this to work on ComfyUI.
I created these visuals with SwarmUI. I didn't use ComfyUI. Swarm is much more practical and faster...
@53rt5355iz right after I sent it I got it working on ForgeNeo, but Swarm would be so much better, thanks for the reply!
Did it ever cross your mind to say WHAT DOESN'T WORK for you? You know, so someone could comment something useful to help?
This is mindblowing, but i wish there was a SFW version, no matter what i do, women are always with their bewbs visible, which is great for NSFW purposes of course so pls keep a NSFW version, but this model is so amazing for other realistic stuff that it's a shame it's so limited to NSFW. If you prompt a guy doing whatever, it makes an incredibly beautiful and realistic image, but if you prompt a woman, bewbs are out (though amazing bewbs btw).
Easily the best Qwen model out there. The NSFW ability is really good and produces great looking images.
You don't really say which Qwen model it's based on? Like 2509, 2512 etc...
2509... The day after I prepared this model, the 2512 model was released. This week, I will do another study with 2512 and publish it by the weekend.
I can't get it to work in comfyui for some reason.
Try changing the clip or VAE models you are using.
It runs on ComfyUI and doesn't edit images in any way. The reference photo remains unchanged. Why?
Because this is a “Qwen Image” model. It is not a “Qwen Image Edit” model. Qwen has released separate models for these two tasks. If you are working on an existing image, use the “Qwen Image Edit” models. If you are creating a new image, you can use my model.