This is the ANIMA version of my NJ Vector Style LoRA 😊

Since this initial LoRA was made by @MrLoreTaker , i had to make a brand new dataset so i used my LoRA plus Hyphoria to generate a bunch of nice pictures.

This turn out a bit different from the original LoRA but it still deliver a nice style, as usual for my ANIMA LoRA, no trigger words, but "Flat Colors" helps if you need to make it extra "vectory".

(My next step is probably Cute IXL in honor of @im_new_to_this, my greatest supporter of this model derivated from NJ Vector).

Special thanks: Some showcase pictures are prompts from @Zlax straight from Strange Thing Mix showcase. Check their work! 💖

PS: merge of this model is authorized as it is a derivative work of an other creator initially.

NB: for the more curious, this one is using a mix of several epochs averaged together using sd_mecha, including a early one to get some craziness back in 😁 (i did not try yet merging stuff with ANIMA checkpoints, but ANIMA LoRAs trained using Kohya sd-scripts works with this tool).

import sd_mecha as sdm
models = [ f"NJ_Vector-{i:06d}.safetensors" for i in [2,8,9,10,11,12,13] ]
models = list(map(sdm.model,models))
sdm.merge(sdm.n_average(*models),output="NJ_Vector-ANIMA.safetensors")

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.