Not loaded
Download
1 variant available
fp16 SafeTensor
BML_Dongsaengz_ANIMA-000019.safetensors
Half precision, best balance • 183.33 MB
Verified: 7 days ago
508
1.9K
License:
AnimaThe Anima Model is licensed by CircleStone Labs LLC. 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.
Built on NVIDIA Cosmos
.jpeg)
Merge versions of the Isegye Idol.
Each version may contain two, three, or more members of the group, and it's still a work in progress.
I realized my limitations when I tried to merge six members into a single LoRA. I don't have enough knowledge about merging, and each dataset contained many of the same tags. Therefore, a LoRA that can represent all six members will not be released anytime soon.
Purpose: To generate images of interactions between members based on the ANIMA model's natural language understanding capabilities without style bias caused by using multiple LoRAs.
Limitations: Costume details are less accurate than in the individual LoRAs. If interactions are not needed, the individual LoRAs are recommended instead.
Goal: Creating merged LoRAs for diverse member combinations. My first priority is Unniz and Dongsaengz, each containing three members. I thought separating the members into these groups would minimize overlapping tags between the datasets. Additional versions are also planned, including Gaenariz, Hentaiz, and costume-themed sets such as SFY_Unniz, SFY_Dongsaengz, and more.
Tips: You can write natural language prompts based on trigger words, like "A girl with grey hair and purple eyes…"