nArnima - personal style for ANIMA

ANIMA V1 is HERE! And nArnima V2 still deliver great results with it: id="update:-i-know-i-have-been-spamming-with-anima-lora-but-i-think-i-have-settled-at-last-on-the-style-i-will-use-moving-forward.-it's-a-merge-of-experimental-and-narnima-v1-(it-includes-te-layers-so-may-not-work-exactly-the-same-on-site).">UPDATE: I know i have been spamming with ANIMA LoRA, but i think i have settled at last on the style i will use moving forward. It's a merge of Experimental and nArnima V1 (it includes TE layers, so, may not work exactly the same on-site).

So, now that i have multiple ANIMA LoRA, for a bounty, i test stacking a few of them: then, i did a quick test on a more "basic" image... and realised it was VERY close to be a style looking like HoJ for ANIMA, including some crazyness 😍

Now, i could have used this stack of LoRA as-is or... merge them as a single LoRA to simplify its usage 😅

So, here i go:

# python merge.py nArnima.safetensors "<lora:CyberDark-ANIMA:0.8>" "<lora:new_mecha_anima_p3-000010:0.6>" "<lora:skzW0rld_ANIMA:0.4>" "<lora:aigoth_merge:0.2>"

100%|████████████████████████████████████████████| 448/448 [07:25<00:00,  1.01it/s]

Saving nArnima.safetensors

Now, i have uploaded it here in order to have my pictures linked to it when i'll use this LoRA. I'll probably also try a full finetune out of curiosity with a model merged with this LoRA later on 😉

Legal

The CircleStone Model is licensed by CircleStone Labs LLC under the CircleStone Non-Commercial License. Copyright CircleStone Labs LLC. IN NO EVENT SHALL CIRCLESTONE LABS LLC BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.