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://civarchive.com/models/790652/uber-realistic-porn-merge-ponyxl-hybrid-or-xl-and-pony-loras-or-controlnet
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.
Liability:
In no event shall I or my team be liable for any claim, damages or other liability, whether in an action of contract, tort or otherwise, arising from, out of or in connection with the use of this model. Please render responsibly.
Description
Merge Recipe:
4 Weighted Sum Merges in total:
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%
FAQ
Comments (22)
I know it's not best practice, but could you add a .ckpt version? Some webUI's don't support safetensors yet
Why can't you use Automatic1111's webui?
some of us are on Macs with M1, invoke and diffusion bee dont support safetensor
I just added a .ckpt version for you guys. Enjoy! Be sure to leave a rating, so that others can easily find this model!
Thanks, will do for sure. Downloading it right now
i like using nmkd local sd ui and the 1.8 update added a safetensor to cpkt converter fyi its light enough to have around for some solid features
@zimzaderk nice, I'll check it out, thanks
Hm, where do I find the .ckpt version?
On the download button, there should be a drop down arrow that allows you to select the type @discount_beats
I am having trouble merging this to my dreambooth models. Resulting images all appear noisy and with a pink/purple hue. I am using A1111 checkpoint merger with the following settings:
model A: urpm
model B: my dreambooth model
model C: sd 1.5
multiplier: 1
add difference
This model is great but it usually gives every girl pubes. Anyone know how to reliably prevent that without destroying sex poses? (negative "pubes" or "pubic hair" tends to crop out the genitals altogether)
negatives usually do the job. If there's a result you like, but it has pussy hair or anything you don't like, send to inpaint and modify all parts that can be improved, or to add jewels, change eyes color, remove pubic hair, etc.
I keep getting a pickling error when using this with google colab stable diffusion webui. How can I fix this?
I read somewhere that it had to do with the filesize. I hope the new pruned version of the model fixes the issue.
Crashes my Stable Diffusion UI.
Loading model from E:\stable-diffusion-ui\models\stable-diffusion\uberRealisticPornMer_urpMv1.ckpt
Traceback (most recent call last):
File "E:\stable-diffusion-ui\ui\sd_internal\runtime.py", line 231, in mk_img
yield from do_mk_img(req)
File "E:\stable-diffusion-ui\ui\sd_internal\runtime.py", line 302, in do_mk_img
load_model_ckpt(ckpt_to_use, device, req.turbo, unet_bs, precision)
File "E:\stable-diffusion-ui\ui\sd_internal\runtime.py", line 94, in load_model_ckpt
sd = load_model_from_config(f"{ckpt_to_use}.ckpt")
File "E:\stable-diffusion-ui\ui\sd_internal\runtime.py", line 636, in load_model_from_config
sd = pl_sd["state_dict"]
KeyError: 'state_dict'
Exception in thread Thread-1:
Traceback (most recent call last):
File "E:\stable-diffusion-ui\ui\sd_internal\runtime.py", line 231, in mk_img
yield from do_mk_img(req)
File "E:\stable-diffusion-ui\ui\sd_internal\runtime.py", line 302, in do_mk_img
load_model_ckpt(ckpt_to_use, device, req.turbo, unet_bs, precision)
File "E:\stable-diffusion-ui\ui\sd_internal\runtime.py", line 94, in load_model_ckpt
sd = load_model_from_config(f"{ckpt_to_use}.ckpt")
File "E:\stable-diffusion-ui\ui\sd_internal\runtime.py", line 636, in load_model_from_config
sd = pl_sd["state_dict"]
KeyError: 'state_dict'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "E:\stable-diffusion-ui\stable-diffusion\env\lib\threading.py", line 932, in bootstrapinner
self.run()
File "E:\stable-diffusion-ui\stable-diffusion\env\lib\threading.py", line 870, in run
self._target(*self._args, **self._kwargs)
File "E:\stable-diffusion-ui\ui\sd_internal\task_manager.py", line 208, in thread_render
for result in res:
File "E:\stable-diffusion-ui\ui\sd_internal\runtime.py", line 238, in mk_img
modelFS.to("cpu")
AttributeError: 'NoneType' object has no attribute 'to'
I've had this happen, too, but only at SD startup. If you ensure that another model loads at startup, then switch to this one, it should be fine. In my case, the crash also corrupts my xformers, so I have to delete and reinstall them if it happens.
Out of ram, almost certainly. Closing all programs including chrome should help. Avoid switching from big model to another big model.
Check with task manager how RAM fills up.
What does it mean when it says
Requires: vae-ft-mse-840000-ema-pruned.ckpt
?
How should I be using the file?
It depends which UI you are using, but if you're using Automatic1111, you do the following:
- Download/copy the VAE here: \stable-diffusion-webui\models\VAE
- Under Settings->User Interface->Quicksettings list, add sd_vae. Mine looks like the following, since I also have a dropdown for hypernetworks:
sd_model_checkpoint, sd_vae, sd_hypernetwork
- Restart Automatic1111.
You should now see more dropdowns to the right of the checkpoint dropdown.
I generally use vae-ft-mse-840000-ema-pruned.ckpt for everything aside from Anime. For Anime I use the AnythingV3 VAE.
The VAE is now included in the model. It is no longer required starting with version 1.2. Hopefully this solves your issue.