Real Dream Website (Faster download)
Available at TensorArt: https://tensor.art/models/899250342552672189/
Comfyui Workflows:
Anima text2image with Loras Workflow
Anima img2img (denoise) with Loras Workflow
Klein 9B Edit with Loras Workflow
Klein 9B img2img (denoise) with Loras Workflow
Z-image-turbo img2img workflow
Z-image text2img with loras workflow
Z-image De-Turbo with loras workflow
This model is capable of generating very realistic images that can confuse people's critical sense, please do not use it irresponsibly, harming ordinary people or public figures. Also avoid its use in electoral campaigns, promoting or damaging the image of candidates. Although the field of artificial intelligence is poorly regulated, its improper use can cause legal penalties even in countries with little regulation in the area. Be very careful to not commit crimes in your country and not to promote injustices against innocent people. I do not authorize the use of this model for Lora training using the face of real people without authorization from the biological owner. Do not create images that can embarrass and damage the public image of people. This can cause serious psychological harm, including deep depression. I do not authorize the use of images generated by this model for scam applications and the dissemination of false information.
During the fine-tuning of AI models, I only use synthetic images that do not infringe any copyright, personal image rights, or laws. The models are merges of open models from different sources and may contain diluted data remnants in the neural network. I am not responsible for possible data remnants since they are diluted, and it is not humanly possible to track the existence and origin of this diluted data.
I do not allow the upload of any of my models on Tensor Art. I only do not allow it there because the policy of duplicate models on Tensor Art prevents me from uploading again if someone has already done it.
I have enabled to automatically hide all NSFW posts for several reasons that are not worth listing. I hope you have patience and understand.
Description
Use with CFG between 1.5 and 2.0
FAQ
Comments (9)
Nice image quality/steps ratio for your latest POny Lighting model, but once again (and maybe i'm just missing the point ?) not really a pony model, barely able to do 'softcore' images, I got like 1 'hardcore' NSFW image out of let's say 20, with a very very hardcore NSFW heavy prompt (that all other pony based model would transform into some obviously very hardcore images) as a test,
I'm not trying to say your model must be NSFW, i'm just curious why you chose the pony models and names if they are not because that is the main strenght of pony models? except this nice image quality again
The model is Pony because most of it is really Pony v6, but as the Pony model is anime style it is very difficult to transform it into a realistic model, just as it was difficult at the beginning for people to train the SD1.5 for anime. The goal of Real Dream models was never to produce NSFW images, the goal of finetune was always to generate more stable images and secondly to generate realistic images. However, in the Pony-based models I made a point of preserving the characteristics of the Pony models and carried out several tests for this. What happens is that at the current stage of realistic Pony-based models it is very difficult to maintain realism and preserve the generation of Pony scene compositions at the same time. In the case of the Lightning models, they are a little more realistic than the standard SDXL and are much lighter, but due to the low CFG and the Lightning Lora itself, it slightly reduces the capabilities coming from the Pony.
@sinatra Do you use NSFW images to train it? It seems to me you would need to in order to teach it what realistic hardcore looks like otherwise you might only be able to do fine art nudes. Maybe I don't know how it works...
If you are having problems with the download on CivitAi, you can download it here on these mirros, but preferably download it from the website to help promote the project.
Guys, you keep using cfg above 2 in models, lcm, turbo, lightning and etc... It's wrong, your images will have strange contrast and bugs. I don't know what to say to you, there are times when it seems like you don't read the description
Heh, xD I don't usually use turbo etc - I just read the instructions now ;) even at CFG 7 though, i enjoyed how bright it was XD I'm a weirdo i know - but I am testing it at low CFG now :)
hey man, what about a turbo 8 version?
Greatest SD 1.5 model I've ever used for now.
Thanks very much, Sinatra. This pony checkpoint is the only really "realistic" one for the moment. All other ones look like ugly 3D rendering. I also love that it doesn't need the stupid string "score_9, score_8_up", etc...
In my opinion, the only flaws are:
- The dull facial expressions. As I like bdsm, I would prefer sad, afraid, or angry faces than smiling or neutral.
- The overloaded backgrounds, even when I just want plain white.
VERY IMPORTANT FOR USER: Don't forget to set CFG between 1 and 2, otherwise the results are much less realistic.