If ever a model deserved for me to say that it is the best I have created, maybe it is this one, what makes this model special is that it is a DMD model made to create super fast images but unlike others, it does not lose quality despite not using Hires.fix, of course you can use it but the results still seem great to me, without going into terms of hardware because I know that we all have different ones, I managed to create sharp images in 2 seconds, using Hires.fix in 6 seconds, I know that some may find details to improve but I really hope that you can try it, it will surprise you.
The VAE is BAKED
Upscaler: Im using 4x-ClearRealityV1 upscale by 1.5 Steps 5 denois 0.2 Hires CFG 1.1,
Width and height: 896*1152
CFG Scale: 1.2
Steps: 4
The sampling method i use is LCM-Karras
It is important to follow the recommended parameters since not all samplers work correctly with DMD models.
This model accepts negative prompts, so for greater realism I recommend:cartoon, illustration, anime, painting, CGI, 3D render, low quality, watermark, logo, label
I truly hope everyone can download it, use it, and share their examples, as this is the only way to continue improving. Greetings to everyone.
Description
With so many quality options available, you need to add something good to the equation to make it worth downloading. In this case, I'm bringing you my 4-step DMD model. What does this mean? I believe I've achieved images that closely mimic reality, but with the added benefit of great consistency and quality, all in just 4 steps. I truly hope you'll try it out and share your comments and examples so we can continue improving it.
The VAE is Baked
Upscaler: Im using 4x-ClearRealityV1 upscale by 1.5 Steps 5 denois 0.01 Hires CFG 1
Width and height: 896*1152
CFG Scale: 1.2
Steps: 4
The sampling method i use is LCM-Karras
FAQ
Comments (4)
Problem with lora. The quality deteriorates significantly when using lora
Getting black frames every gen
UPDATE: removed clip set last layer, works now
@Sateluco I believe it would be good if you reuse the prompts from
https://civitai.com/images/117918757
just to compare DMD with non-DMD.
As you might remember I argued that DMD destroys the details and composition richness. Maybe it is not true already?
Man, that's crazy. I've been through all the SDXL checkpoints, the big ones and many of the smaller ones too—but the level of detail your checkpoint provides is out of this world. Insane!