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 (19)
@bilibilissr I am not sure what you mean by "repair model", so I'm gonna assume you meant inpaint. The inpaint model is used for img2img inpainting inside of the a1111 webui. Inpainting is when you put an already done image and you draw over it, making Stable diffusion redo the part you drew over with what you put in the prompt. For example, in the showcase images, I drew over the girls bikini, and typed in "red bikini". Therefore, Stable diffusion changed it to red. This can also have other uses as changing hair, the color of eyes and the clothes. And you can also add new objects to the scene. It basically let's you edit anything into the image.
@Manseo @Manseo Translation: Thanks for the reply, because I have come across models specifically used to repair/modify image content before, but I am not sure if there is such a specific model, it is too rare. Then I saw the name of your model "v2.2-inpainting", so I wanted to know if this model is also a model specially used for repair. Now I feel that I must have misunderstood, and I am covering my face.
@bilibilissr The reason you don't see inpainting models very often is because people usually can't be bothered to make them. Inpainting is more of a niche feature of stable diffusion. Plus, you need to make a different model for it to work, so it's just extra effort. If somebody hadn't asked for an inpainting version, it's most likely that I would never have done it as well.
Tried with a few self made LoRAs and the results were exactly what I was looking for from quite the time.
I am always grateful to you. Of all your models, I love the 1.7 the most. If I put version 1.7 and the latest version in one folder, will the latest version be applied even when I use version 1.7?
@etrange1004 Thanks for the support. The version that will be applied will be the one you've selected in the webui
Does any version take anthro/furry prompts?
@Omikonz Sorry for the late response, but yes it can do furry prompts. I tested on the latest version and it did give some pretty decent results. Although you need to put "(furry:1.6),(anthro:1.4)" into the prompt.
请问这个大模型可以用来训练人物lora吗？需要自己手动添加VAE吗？
great
作者你好，我很喜欢这个模型，考虑弄个XL版本的吗？
I love this model and my go to to use. Any plans for a XL version?
@zenrel Hi, thanks for the support. I am busy with life at the moment, but I think I will start working on the XL version in summer.
Quite a decent cute flat anime model. Thanks for pointing out which models were merged, not all of them do it. It's a good choice, so this merge turned out to be very good.