Daily Release Campaign has ended!
See here for more info: URPM is back - Daily Releases incoming!
💖Join the Patreon if you want to support our work & even earlier access to models being beta or alpha tested!
🗣️Join the URPM Discord for Updates or Support!
---
My SDXL and Pony-SDXL Hybrid models are now available here: https://civitai.com/models/790652/uber-realistic-????-merge-ponyxl-hybrid-or-xl-and-pony-loras-or-controlnet
Additional Downloads:
When clicking the arrow to the right of the primary download, you can choose between pruned and unpruned versions. I recommend the unpruned version when merging with other models. You can also find the inpainting model as a separate version below.
Example Images:
None of the sample images were altered, upscaled, etc. All can be reproduced using the metadata via the unpruned model (however the pruned model should now have the same results).
Prompting help:
If you notice that it’s not doing what it should do, be extremely light with the negative prompt. And then I re-use the same seed and add more words when needed.
Just like any NSFW merge that contains Stable Diffusion data, it is important to use negatives to avoid combining people of all ages with NSFW. This is sadly unavoidable without adding negative prompts, until there is an embedding or the like that can help automate this process. Here are a few things that I generally do to avoid such imagery:
I avoid using the term "girl" or "boy" in the positive prompt and instead opt for "woman" or "man". The only exception is when I am specifying how many women should appear in a scene, which I would then use "1girl" or "2girls".
In the negative prompt I use: child, childish.
This has helped me prevent any kind of accidental imagery. I know a lot of us are used to using the term "girl" for "women", but AI can't understand the difference.
Description
Pruned the model. Reduced the size to 1.6gb from 8gb. Not bad. (Unpruned Version also available (best used for merging))
Rather than using Weighted Sum Merges, I instead merged via Add Difference with the Model A always being used as the tertiary model. This allowed always removing any duplicate data that existed in other merges I was merging with.
By doing the above, it actually fixed problems more frequently in regards to fused limbs, wrong limbs, etc.
Merge Recipe:4 Add Difference merges in total (with Model A always in the Tertiary position as well):
1st Merge:
izumi
sxd-berrymix-merge at 35%2nd Merge:
^above
ZombiMix-v7 at 15%3rd Merge:
^above
3DKX_1.0b at 15%4th Merge:
^above
RealEldenApocalypse_AnalogSexKnoll_4CandyPureSimp+FEET at 25%