Try it out here! https://mobians.ai/
Join the discord for updates, share generated-images, just want to chat or if you want to contribute to helping perfect the model! https://discord.gg/KfAuP72uCd
USE:
The model uses Danbooru/E621 tags similar to anime models. I would not recommend just generating a character with no follow-up prompts as the results won't look very good. A good negative prompt also helps greatly.
Limitations:
Some characters generate less consistently. This problem will be consistently worked on as new versions progress.
Description
FAQ
Comments (55)
I wonder if a SDXL model is in the works? Mind you, this model is fantastic for being 1.5. I was using SDXL models exclusively for other things but this seems to be the only really good Sonic characters SD model.
I tried making one at some point but it wasnt very good. Honestly going to wait a while for the ecosystem to mature. Its too expensive to rent GPU's to experiment with.
@metal079 Understandable. Main reason for SDXL for me would be the many artistic Lora’s out there.
Nevertheless your model here is great, I’m having a lot of fun generating Sally art with it! 😅
I tried the Mobians.ai website and this new V4 update seems noticably more stable than the previous ones, good job. Whatever it is you are doing seems to be working.
Love the model Metalll (-Koi)
I can't seem to replicate the results from Mobians.ai anywhere else. What settings does the site use?
Does it do Sonic Prime characters or no?
For when V5 and that it can run well and without problems in qDiffusion?
fp16 version when?
Every time I try to generate something via Stability Matrix, I just get a grey square as my final result instead.
Same and it irritates me
Why do the girls have insane abs
Civitai is testing a new generation backend, and unfortunately generates more black or gray squares than actual images...
Tell me about it Bro, what are they testing man?
Yeah, having the same problem...
Yup
Same problem here.
It's weirder that it seems to be only model from one's I've been using to have this problem, rest work - MOST of a time.
@OmegaArts13156 You can tell when the problem is on the website's part, and when it's the model's fault
I've not been able to generate anything but black or gray squares at all today.
@Rikmach ...Yeaaaah it's been like that for few weeks now.
Yeahhhhhhh.... I don't get it either... I wonder if it's broken and they are trying to either fix or replace the model tensors....
@LeoTheLegend76 idk... At this rate the answer is meth
hay! For some weird reason, most of the time when I use this model, it just gives black, gray, or horribly pixelated results. mind figuring out what the problem is?
I don't know but I hope it gets fixed
adding a VAE seems to fix it
@blckroughdmnd560 which do you use?
@fallendragonwolf115 I stuck to ClearVAE but any of them should work
When will you upgrade to the SDXL version?
Its in beta rn
Can you make a Pony based XL version?
I very badly want this too.
Black and Grey images appear for constantly regardless of Lora or vae inserted ....this pains me in more than one way
You're not the only one that had problem
If you're using A1111/Webui-Forge i set the "SD VAE" to"Automatic" and i got clear results instantly. No Black or Grey screens. Hopefully this helps. (Source: I struggled for hours to get this to work lol)
@DidntiSaySo what about in comfy ui?
As an artist myself, I like to use AI more as inspiration. While others say AI is toxic, I see it more as an advantage point. It's really good at helping me come up with inspiration for art and character designs. It's really good, honestly.
I feel like ai art has huge potential to be used as a ref that artist can use or like you said used for inspiration. Many artists are against it which is understandable but at the same time it can be used as a great tool help beginner artists and experienced artists with drawing things that could be difficult to find a ref for. I personally have had a hard time finding good refs but i feel that ai art has helped me a decent amount with that. Like I said i believe that ai art should be seen as another tool in an artists arsenal rather than the black plague lmao.
i honestly should've kept my furry filters on, i just wanted normal sonic pictures 💀
Wish i could download an exe of it even if it had limits
nice
I saw this in another comment, but I too (when using the model) have it generate within Black and white more often than not.. even after playing around with different prompts/negative prompts. Jump bumping back up to the top of your radar.
I'm just here for buzz bro help my ass ;-;
any remaster ?
I'm sure someone has asked already, but is it possible to get characters like Whisper or Infinite to generate with their masks on?
How come I can't use this model anymore? What happened? This is my favorite model!
My images aint looking like that what do i do?
Why can't we use it anymore? Why do we have to bid?
Because Visa and Mastercard are trying to hide their closet secrets by forcing Civitai to remove what made people stick around and use the site.
oh.
not work for me still cant get to on my stable diffusion to fix grey image? any help
What happened? 😿😿🤧🤧
Hi I'm trying to create a pure white Whisper, but I need her mask to be a crucial part of her. I'd love to hear your thoughts
"Error Message
There was an error attempting to generate your image. Website is possibly down. Try out generating on CivitAi as an alternative. https://civitai.com/models/1493"
Are there're intentions of reviving Mobians.ai?
I see no probabilities of this model being elected to be used in CivitAI's generator.
Details
Files
Available On (2 platforms)
Same model published on other platforms. May have additional downloads or version variants.