Download
1 variant available
The 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
Trained on 32 images of mvp ends and images from the gallery [03.2026]
For Anima version 1: it is best to keep the power of Laro at 1.5 for the best results. Using additional style hints will critical change the default style
For Anima3 version 2: Here I tried to increase the weight of the model, this improved the consistency of the style, and also solved the problem of the hands that I had on every model. As an Upscale, I used HiresFix on 1.5 with a 4x_foolhardy_Remacri or RealESRGAN_x4plus_anime_6B model, 14 steps, and 0.4 strength. Unlike the previous version, 1 strength is sufficient for all scenarios.