License
Because of the base models this model inherits some terrible licenses 😔
Dreamlike Diffusion 1.0 License
Seek Art License
PROMPTING
The model is easy to prompt for, words like blurry background, depth of field are good for example:
Positive Prompt:
1girl, asian, blurry background, depth of field
Negative Prompt:
worst quality, lowres, ugly, tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, extra limbs, disfigured, deformed, body out of frame, bad anatomy, watermark, grainy, signature, cut off,
I usually remove blurry or blurred from the negative prompt
Words like film, and photo are good as well:
Positive Prompt:
1girl, film, photo, depth of field, gradient hair,
Negative Prompt:
EasyNegative,, worst quality, lowres, ugly, tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, extra limbs, disfigured, deformed, body out of frame, bad anatomy, watermark, grainy, signature, cut off,
I also removed blurry or blurred from the negative prompt for this image
Because of the realism of this model I used it to replace the person in this video with emma watson using TikTokDiffusion you can see a side by side comparison
Description
FAQ
Comments (5)
"This is not a photo" is a strange way to begin the page, as we've all seen photorealistic models before and so we aren't surprised by it. Would make sense if you were aiming the page at people not familiar with stable diffusion, but such a person wouldn't be here.
100% agree.
I’ve scrolled by this model so many times today, just now realized it wasn’t an advertisement hahaha.
Initial thought after realizing it wasn’t an ad was “how would some release a video based model? Doesn’t make sense.”
You're fun at parties. All of them.
If possible, can you add the smaller 2gb pruned-fp16 version ?
The newest version is smaller
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.