Not a big thing. But people asked for it so I'll upload it here for everyone to use.
Bases are:
AbyssOrangeMix2 (hard) (75%)
Lawlas'sYiffyMix (25%) - https://civarchive.com/models/4698
Discord Channel for many examples by the community:
https://discord.gg/furrydiffusion > Custom Models > Lucid's Mix [Model]
Common Settings:
Steps: 20-25
Sampler: DPM++ 2M Karras / Euler a
CFG Scale ~7
Description
Same model but saving a lot of storagespace
FAQ
Comments (29)
Looks nice, but the file is too big, please shrink it with toolkit : https://github.com/arenatemp/stable-diffusion-webui-model-toolkit
Right now my webui isn't booting anymore somehow. But as soon as I can get it working again, I'll replace/add the shrinked data!
Is this a difference merge or a sum merge?
oh and where can i get the michiru lora
What exactly does that mean? I'm sorry, still pretty new!
@Fonfurn It's by a friend who is offering it for download on a Furry-AI-Discord. I'll post the link in case you want to join!
https://discord.com/channels/1019133813105905664/1072911691907596308
@LucidAI https://www.reddit.com/r/StableDiffusion/comments/z9wbwk/what_does_weighted_sum_and_add_difference_mean_in/ also thank you you are a savior for sending that discord link
upon realization i cannot join there
@Fonfurn Well, here's the direct download: https://mega.nz/folder/06ZihYDC#10wBgSQr8dg536WoYq9adg
@LucidAI gnarly
but is the furry ai discord private or something cause google is like "boy what the hell boy what is this"
@Fonfurn No, non it's like a public server. I found it via reddit:
https://www.reddit.com/r/FurAI/comments/xz5jva/furry_stable_diffusion_setup_guide_model_downloads/
@Fonfurn There's one on this website with really good results. Could even be that one that was posted in the Discord given the time frame they might have decided to put it public
@Fonfurn Thank you very much, I look forward to the future updates of Michiru
@Natrox The creator is pretty active on Discord. I'll send you a link to his model's Channel. If you got Discord, you can have a look on it. There are also a lot of example pictures and it's prompts/settings!
https://discord.com/channels/1019133813105905664/1072911691907596308
@LucidAI can't open discord link
Where do I find the hypernetwork you used in the examples?
you gotta look for the torrented NAI file on rentry
What is this "hypernet framework" in the example prompts? I wanna make sure I have everything.
The Hypernetwork should be " furry_kemono " for the most ones!
@LucidAI Can you share a download address with me? I looked for a long time but couldn't find it
@tootwo Unfortunately this Hypernetwork is not exactly something OP, or I can share, as it was apart of the leaked NAI Diffusion Bundle, and the team behind NAI is EXTREMELY against free distribution of the community assets they sell a subscription to use, I would suggest you look into finding the original leak, but I must ask you not to do so, as Pickle Tensors are vulnerable to code infection, and there's no guarantee copies floating around aren't infected with malware. As it stands now that's all I can say, sorry for the bad news.
@tootwo
https://file.io/Xh0RFs4Jz4TQ
Up to you if you trust it, run it through a pickle scanner and ask someone else that has it if the hashes check out.
It looked nothing like the example pictures for me sadly
That's really unlucky. Keep on it and maybe try different LoRAs and VAEs. If you need any help, hit me up! - Best bet is to join the Discord I linked. There you can find a big community and many really skilled people using that model. Those are creating much better stuff, than I could ever imagine. Hope you'll find a way to appreciate the model!
I don't know if you use xformers, but if you do then know that it causes non-deterministic results as a side-effect of how it works. Essentially, you won't ever get the same image from the same seed and settings. If not, the image you're looking at could be using hypernetworks or other addons/extensions to further tweak the output. If you don't have those, then your results will differ as well. It's also been noted that having a different GPU can affect the generation results as well. There's a lot of factors at play, just a friendly heads-up.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.