Sweet-mix
End of the road - After 22 versions and over 2 years, Sweet mix will no longer be updated as of Illust v1.4. Honest thanks to those who stuck around.
Sweet-mix is a text-to-image Stable Diffusion model that is mostly focused on the anime art style. The model itself is a merge of multiple other models, as i don't have neither the time, nor the hardware to train a model. Sweet-mix can be used for both SFW and NSFW images. I think my model is pretty nice, so feel free to give it a try.
MUST-HAVES FOR GREAT IMAGE GENERATION LIKE IN THE IMAGES:
These quality tags: masterpiece, best quality, high quality, detailed, absurdres
This negative prompt: ugly, bad feet, bad hands, bad art, ugly artstyle, bad anatomy, bad fingers, censor, censored, ugly, deformed, noisy, blurry, low contrast, photo, realistic ,watermark, username, text
VAE: SD XL VAE
Good artist styles to use: iwbitu, kesoshirou, shexyo, oroborus, mamiyama, kihou no gotoku dmc, seisui bar, kawase seiki, yuuji (and), shiokonbu, hitmanatee, ashraely, starmilk
And a great imagination :)
Tensor.art link: Sweet-mix - Illustrious/XL - v1.4 | Tensor.Art (free online generation)
Description
FAQ
Comments (9)
That "(score_4,score_3,score_2,score_1:1.3)" negative prompt is pointless because pony doesn't even have it. It's rated between 9 and 4
Hi, thanks for telling me. I'll change it in the future
I've done a bit of testing and I noticed that if i remove the "useless" parts of the prompt I get worse results. You can remove them if you want to, but I'll keep the negative prompt the way it is now for the time being. If it ain't broke don't fix it.
@Manseo "Oh, and one more important note - these tags are less useful in negatives as you can only go as low as 4 and ideally we should train on quality all the way down to 1, but it makes the training significantly more expensive. So you can still use score_4/5 in negative but they will not push you away from really bad images." from author of pony from his article about scores
@DevilLady As i said, you can remove them if you want to, but i am keeping the negative prompt the way it is for the time being. I tested without them and i got worse results.
@DevilLady my guess is the model may have extrapolated that lower scores are stinkers from the fact that 1 is less than 4.
A very good model, you should integrate VAE into checkpoint, it will be much better !!
ayuda soy nuevo, la unica forma de ejecutar este modelo es con difusion? y que plan recomiendan comprar
Hola, no hablo español, así que estoy usando el Traductor de Google. La forma principal de ejecutar modelos como estos es ejecutándolos en tu computadora localmente. Puedes hacer esto si tienes una buena tarjeta gráfica como una RTX 4060. Si tu computadora no es lo suficientemente buena para ejecutar Stable Diffusion, tengo una alternativa gratuita vinculada en esta página llamada "Tensor.art". Te permite ejecutar el modelo de forma gratuita en línea