Cute Style Mix

Cute Character Style Mix

This is a blend and merge of some various art styles and some of my previous Loras, meant to capture a cute/chibi character aesthetic.

Version 1 does include a merge of my Care Bears Lora, https://civitai.com/models/396604/classic-care-bears-pony and

Berry Brite v1 Lora, https://civitai.com/models/369367?modelVersionId=412745 as such if you wish to use either of those Loras with this one you will have to lower the weights of one or both, or it will give overbaked results. Using the v2 of the Berry Brite Lora does also have a little overbaking (as it was trained off the v1), but it's less apparent.

Version 2, 3 and 4 only include the Berry Brite Lora in the merge, so the Care Bears Lora should work fine with them.

Because it includes these Loras in the merge, it's able to use the trigger words from them to give different outputs, however it doesn't capture the styles of those two Loras.

First example is a base set, with no trigger words,

v1

v2

v3

v4

next is using the CrBr trigger from the Care Bears Lora, (only merged with v1),

As you can see, the differences are minor, mostly with the shading. The base outputs seem to have slightly sharper edges to the shadows. Also some animals will look different (especially those that have actual care bear character counterparts, such as the rabbit). This model also has a higher chance of giving anthro/animal character outputs without specific human prompts.

This is less of a case with the next set, the tkid and tkd triggers from the Berry Brite v1 Lora,

first with tkid,

v1

v2

v3

v4

then tkd,

v1

v2

v3

v4

You may get more human characters with these tags, but I'd still suggest prompting for them in any case. Version 4 leans heavily into anthro characters, even with these tags.

Now there's the rbrite and rbrt triggers, which go back to more anthro/animal styles, and are nothing like their Rainbow Brite inspiration,

with rbrite,

v1

v2

v3

v4

and rbrt,

v1

v2

v3

v4

Then final two triggers sberry and sbry were the two I couldn't get to work right in the Berry Brite v1 model, but they still gave interesting results there, and that is the case here as well. Sberry may give more human outputs, while sbry will likely give more anthro. (As stated above, v4 tends to give mostly anthro outputs with any of these tags).

sberry,

v1

v2

v3

v4

and sbry,

v1

v2

v3

v4

Overall the style changes for Version 1 are minor for the old trigger words, while Version 2, 3 and 4 have more variety. But I felt like showing these examples just to give people an idea of other possible prompts/triggers you could use to get different results.

I know this Lora is a bit all over the place, but I hope some of you can find a use for it.

I don't have any plans to make more versions of this model at this time.

All of my models may be used for other training and merges.