Download
1 variant available
82
173
1.3K
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
Model trained on Anima base-v1.0 to imitate the style of MoNo
=====
Support the artist: https://www.pixiv.net/en/users/93740696
=====
Anima seems fascinating, so I wanted to give it a test by training the dataset I was happiest with on NoobAI. The results were... surprisingly very good despite no changes to the dataset from my end and just using a training script I found in the comments section of the Anima model's CivitAI post without even touching the settings (That script being Anima TrainFlow by ThetaCursed, which can be found here: https://github.com/ThetaCursed/Anima-TrainFlow). This one is even closer to the artist's style than either version of the LORA I trained on it for Noob, and I'm sure if I learned even more about making the process more finetuned it could be improved even further...
NOTE: All the images I generate for the showcase are done with base Anima without the use of any quality tags, Adetailer, or Inpainting, only Hires. fix to do very basic upscaling. Anima is a base checkpoint and not too many finetunes exist yet, so I haven't really tested the LORA with those at all, but the results speak for themselves on the base.