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
FP8 Version: Use Forge or ConfyUi
This version is the final version after 38 beta versions. This may be the last version based on the SDXL architecture, fine-tuning is getting difficult.
FAQ
Comments (24)
I see no big difference between this and V12. But that homeless crackhead SpiderGwen.
Nothing seems to have changed, but the faces are grasping better, as well as the detail has increased. Thank you so much for the new model
everything i use looks like an old cga monitor with very blocky bright colors. What settings do I use?
Using a VAE?
after doing an accidental run on a low cfg, it looks good, well its making it anyway. I found cfg 1.5 to be the sweet spot, though it doesn't really respond to instruction.. I don't have any vae... Is there one baked in or is that a part of the problem?
But its weird anything higher than 1.5 and everything either looks like solid cartoons, or just really bright garish colors like really old computer graphics from there 1990s. But at least I can play with it now.
@frankmike DMD models require 1-1.5 CFG
@frankmike You downloaded the wrong model.
@tymirrainier865 I guess so... There are so many variations to these models its hard to keep track of them all. Though that dmd tip I have to write down. So far its working at 1.5 but the creativity really lacks and the faces all look the same.
that said i did download the SDXL plain model.
are there any ideas how to attenuate the "pony face" look? it's hard to describe, but pony seems to make, for example, the outlines of the jaw a certain distinctive way. i mean pony models in general, not picking on Real Dream.
Is it puffy, i dont see anything else. Better, than flux chin, which is recognizable from miles.
Maybe try a character LoRA? Found that using a Pony or SDXL LoRA at low strength (usually ~0.3 to 0.4) is just enough to tweak the facial appearance without making it actually look 100% like the intended character. Unfortunately, some Pony models react differently to this more or less than others.
@fullersid820 yes seems a issue with any trained SD model puffy face narrower eyes than training model. Never got a really close match with making a lora vs. training model.
Great model, any plans on doing Illustrious?
FP8 version available
are fp8 just faster gens what is the difference between fp16?
@JamosBond 4gb instead of 6. I dont know, if its faster. I have 8gb card, and i use the fp32 clips for Pony, with same speed.
Thank you. I wish more people would post FP8
absolutly fantastic, the best i've tried so far
Is anyone else having this issue? Real Dream Pony doesnt seem to generate coherent faces for me. Most faces are off.
same here...faces look way off
@seath1337 the new model fixed it significantly for me!
sdxl pony 11 is insane, it creates images similar to Flux, with the right amount of soft textures and sharp outlines at low CFG, even using it as an adetailer model creates a face better than any of the other models I have. It may start to struggle with hands the longer your prompt or the more Loras you use but damn this is close to the best for character creation
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.