PLEASE READ DESCRIPTION
Deprecated model, consider updating to: https://civarchive.com/models/147363/requiem it is also licenseless
Also, we now have an sponsor!! You can use Liberty model at their website without any hardware or installation requirements. You can also create fun animation on Mage with my mode!: https://www.mage.space/
Extremely NSFW biased model!!! But awesome at SFW too. Use it at your own risk of getting unprompted explicit images.
This is not an easy to prompt model, nor the best one for begginers.
According to one user: "If you have the model-keyword extension make sure you uncheck it. It appends all the triggers words and results in complete nonsense results. Was getting super frustrated that even copy/pasting prompts gave me trash".
Check any info or questions at our private Discord here: https://discord.gg/z88HpDwbGq
MODEL (there are two possibilities):
Liberty-Main: You should be using this model. It might not be the easiest one to prompt depending on your style. But you'll get used to it soon and is the most powerful one.
Liberty-BadClip: This version uses a broken CLIP model in the same style aEros CLIP model was broken, so outputs are very different from main version. If you really know what you are doing, or you really don't want to change your prompting style from aEros and it performs bad in main version, or are getting general bad results with main one, etc... you could try this one. Keep in mind that some UIs have problems with broken CLIPS. Also that I won't continue or support this version of the model.
VERSIONS (both models have 3 versions + 1 pix2pix version):
Standart: Intended for general use. It has VAE vae-ft-mse-840000-ema-pruuned.vae.pt baked in. There's no need to use any additional file.
Inpainting: Intended to use in the inpainting section of the img2img tab, it is mixed with original SD 1.5 inpainting model and is much more coherent whith masking areas. Only select it for that purpose. There's no need to use any additional file.
Pix2Pix: Main Version exclusive. Intended to be used as a pix2pix option of the img2img tab, it is mixed with original 1.5 pix2pix model. Only select it for that purpose. There's no need to use any additional file.
Training: Intended to use in the Train tab, for embedding creation. It uses SD 1.5 original VAE baked in, so it doesn't deepfries embeddings. Only select it for that purpose. There's no need to use any additional file.
All of the files are provided in ckpt and safetensors format for your convenience ;)
ABOUT:
Liberty is freedom. A merge with over 23 other models with a methodical, careful and genuine approach. Check 'CREDITS' section for the full list.
Freedom of prompting art or photo or both, landscapes or backgrounds or interiors, people or entities or scenes, stiff poses or movement or even mouth and facial emotions, SFW or nudes or even hardcore sex. I tried to make it as versatile as possible and merged it with half CivitAI to get the most out of any free model out there.
And it is free to use for any open source purpose, commercial or not. All the models I used were licenseless when I grabbed them, although some have changed licenses afterwards. However it is also much more modular in it's development process, which means that if any problem arised, I could rebuild it much quickier to avoid getting chained to a license or other type of problems.
HOW TO USE:
This model does not need to use any of it's trigger words. They are a tool, and the knowledge of those trigger words is deeper known and better accesed directly. So 'a photo of a cyborg woman' will work better than 'cyborgdiffusion, photo of a woman', but you can also try 'cyborgdiffusion, photo of a cyborg woman' or even 'cyborgdiffusion, photo of a knight woman' for special and unique effects. If you want to get an idea of what they can help to achieve visit the 'CREDITS' section to check the models they come from.
I mainly try natural language for my tests. You have an example prompt screenshot avaible to build from there if you want to. There's a more advanced prompting guide at my Ko-Fi: https://ko-fi.com/promptmarinersai
It is merged with UnstablePhotoreal (but with much less weight than aEros), and it reacts well to their prompting language, you don't need to use it as comma separated tags: https://docs.google.com/document/d/1-DDIHVbsYfynTp_rsKLu4b2tSQgxtO5F6pNsNla12k0/edit
As with my other models I have tested most intensively how good it is at making erotic and artistic nudes. But this time I have tried many other types of prompts too, some of them borrowed. But keep in mind that some uses are edge case. For example it is not so good at hardcore sex scenes as far as I know, you'll get more consistent and effective sex scenes at other specialised models instead of Liberty. And it won't make the best anime art out there either.
It may not be the easiest model to use, give it some time to get used to it :)
CREDITS (Many thanks in no particular order to all of the authors of this awesome models, part of Liberty, without them this wouldn't exist):
UBER v1.2: https://civarchive.com/models/2661/uber-realistic-porn-merge-urpm
Analog Diffusion: https://civarchive.com/models/1265/analog-diffusion
Artificial Journey: https://civarchive.com/models/5279/artificialjourney-v10-768
Cyborg Diffusion: https://civarchive.com/models/1365/cyborg-diffusion
David Tennant: https://civarchive.com/models/4125/david-tennant
Elden Ring Style V3: https://civarchive.com/models/5/elden-ring-style
GTM V1: https://civarchive.com/models/4525/galaxytimemachines-gtmultimateblendv1
Hassan V1.4: https://civarchive.com/models/1173/hassanblend-1512-and-previous-versions
Knollingcase: https://civarchive.com/models/1092/knollingcase
Myztery V2: https://civarchive.com/models/3947/myztery-a-class-based-fantasy-model
Postapocalypse: https://civarchive.com/models/1136/postapocalypse
Sci-Fi Diffusion: https://civarchive.com/models/4404/sci-fi-diffusion-v10
SynthwavePunk V2 (do not mistake it with the licensed one): https://civarchive.com/models/1102/synthwavepunk
Project Unreal Engine 5: https://civarchive.com/models/4752/project-unreal-engine-5
Unstable PhotoReal V0.5: https://civarchive.com/models/3753/unstablephotorealv5
Vintedois: https://huggingface.co/22h/vintedois-diffusion-v0-1
Wavyfusion: https://civarchive.com/models/1196/wavyfusion
AltroticArt's Penis Model: https://civarchive.com/models/1245/airoticarts-penis-model
Homoerotic V2: https://civarchive.com/models/1256/homoerotic
SimpMaker 3K1: https://civarchive.com/models/1258/aloeveras-simpmaker-3k-series
Let's all have fun and play together!
Description
This is the version you should be using for pix2pix img2img:
VAE is included in the model.
Remember de filename must finish in instruct-pix2pix
FAQ
Comments (19)
Would you consider mixing in smirkingface? There's a sd1.5 comaptible version and one with a different layer structure. https://www.patreon.com/posts/released-models-73689798
I would have to try it get familiar with it and know it's license. But noted for later checking, thanks!
不熟悉英语的人表示“学你们的自然语言比学AI还难”
"People who are not familiar with English said that "learning your natural language is more difficult than learning AI"
When he says "try natural language" he's not saying "ENGLISH IS THE NATURAL LANGUAGE", he's saying not to prompt poorly and try to use natural sentences.
As in,
"A photo of a beautiful girl sitting in a messy room" - natural
VS
"beautiful girl, sitting, messy room, photo" - not natural
etc.
So your 1-star review based on a misunderstanding really isn't fair.
learn before complain
This is the best!!!!
Better at NSFW than 98% other models which claim to be NSFW
100% agree. RealisticVision2 is the only other that comes close for me. It's insane how good they both get when you find the sweet spot in the settings.
Seems Liberty needs low CFG though while RV2 requires a higher value to shine.
@thisisthewaytosalvation Right now I'm using 8.5 CFG for Liberty, I have used it with 11 too.
Where is the yaml config for inpainting?
As far as I know there's not such a thing for 1.5 based models... They inpainting model is just a merge with the SD 1.5 inpainting model. Anyway it works well without anything else
Interesting inclusion of Homoerotic v2, how did you mix it in and what changes did you notice? Always happy to see what people are up to with it, when I find them in the wild
Hi! It was included to try to improve male anatomy, in the first stages it was very helpful in maintaining decent penis anatomy, as I had an original intention of making Liberty capable of good hardcore porn. However as the merge progressed, since I valued much more having amazing and artistic women I had to give up. However, thanks to homoerotic and some other models, males are also amazing in liberty (except their genitals which are most of the time vaginas xD)
@aine_captain lol I understand, glad to hear it! Maybe v3 can help tackle that problem better, fingers crossed
@aine_captain Sorry to say, but 'amazing' might be a bit of an overstatement here. Your model ended up with a light case of 'parts better than the whole', when it comes to men. Analog Diffusion, Hassan, Homoerotic - all produce a lot better photo realistic images of men using same prompt. Your model seems to produce rendered art looking men out of the blue even when asked for a photo.
You didn't include any 'male' tags for your model, so I will avoid rating it. But men do seem like an afterthought.
@SomethingSomethingUsername Meanwhile, I'm making stock photos of males with Liberty. Which doesn't means that Hassan, Analog Diffusion and obviously Homoerotic cannot be better at males than Liberty. I don't know if they are.
And sure, males WERE an afterthought in the model, I only tried to care for them with the idea of being able to make sex scenes at the begining. When that showed not possible if I also wanted the model to be able to do all the other things I was more interested at I switched and cared about males being photorealistic and artistic when not naked. So for nude males, sure, the model sucks. But otherwise I hold to what I said. The model is amazing at males so long you don't want to see their penis.
One quick note, testing a same prompt across different models and arriving at conclusions from it is not the best to go. Models react different to different prompts. I can create any prompt that would be amazing at Liberty and would suck at Hassan's or Homoerotic. I wouldn't say those models are bad at the task just with that limited testing.
But with the right prompting Liberty makes amazing photorealistic males. Yet, I don't claim it being any better than any of the models you mentioned at that task. IDK|IDC
@aine_captain Sorry if I used a bit too harsh of language. I didn't mean to say that your model cannot generate photorealistic men. I meant, that I had inconsistent results - when prompting for a photo, result could change to digital render depending on the content I asked for. Might just a byproduct of this model not being purely photo, but rather mixed.
And yes, you're right, men generate not any less good looking than any other model.
@SomethingSomethingUsername No worries, yeah it sounded a bit harsh, but I've dealed with much worse people here, so I may have reacted a bit too :)
Yes, this model can make very computer like images when asked for photos both for men and women. Try a negative consisting ONLY in this two words and check if it fixes it for you: "grotesque render". (No commas or anyrhing else).
Only good for static poses.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.