This model can generate images at 1600x704 with no highres upscaling required.
Use with Automatic1111 or InvokeAi on RunDiffusion where you can generate your favorite CIVITAI models.
This merge was created to achieve clear and unrivaled image quality, color and sharpness; no more blurry or foggy filter over your image.
Join Civitai-Crystal beta on discord!
EXTRA:
Most of the images are rendered at 768x1216 with NO HighRes Optimization and NO Upscaling of any sort. The faces were not adjusted in any way. These images are raw results. As you can see, the model handles the prompt very well even at large sizes.
VERSION 1b:
In overly colorful pictures some details get lost in the intensity. Toned down the color intensity a little for a clearer and crisper image. Also, there are now extra details in your renders when compared to the previous version of Crystal-Clear.
Try your LORAs through this model and post your results.
Feel free to compare with other models.
Don't forget to post your results!
Description
Pruned version now available.
In overly colorful pictures some details get lost in the intensity. Toned down the color intensity a little for a clearer and crisper image. Also, there are now extra details in your renders when compared to the previous version of Crystal-Clear.
FAQ
Comments (18)
Can this model do photo realism? and is there a pruned version?
I suppose it can, but i have not tried so I will not bet on it. Please try it out and let me know how I should improve.
About the pruned version; in the near future.
im asking if it can cause almost 6gig download ;) is rough
@hetaneko I might need your help with this since I am not familiar with it. Add me on discord and let's converse.
@WarAnakin you can convert any model using this extension https://github.com/Akegarasu/sd-webui-model-converter you have to check the checkbox "No-ema" then model will be pruned to 2gb.
@hetaneko oh man i can't thank you enough. I see you made a pruned version, thank you!
@DukeNukem47 Now you have access to a pruned version :)
@DukeNukem47 I'm not saying more gig's equal's a good model but get used to high memory cost if you want high quality images.
@tinyduck221230 not much changes if you remove ema from the model. Ema is used for training not for inference.
@hetaneko thank you, just got the notif now
Very nice model! Having a lot of fun playing with it!
Thank you! I am delighted you are satisfied with the results. I, for one, am impressed with the renders you achieved. Great work!
I'm looking for beta testers for the next release, contact me if you are interested on Discord :)
Try making one with Tyrande Whisperwind.
@WarAnakin Hey! Sorry I'm not really interested in beta testing, I'm only doing this as a fun hobby I don't really know anything so I'm not sure I actually would have any interesting feed back. Happy to wait for the next release, tho!
Also apparently anytime I make a gallery on another post using images made with this model it adds them to another gallery here also... Don't mean to flood, I don't know how to stop that from ocuring, sorry!
@RoiD1kan It's all good, no worries.
a realy good model, perfect prompting. Its really worth.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.