Model (training image captioning model) minicpmv4.5-en(natural language & English) Parameter Settings Network Module LoRA Use Base Model Anima - preview-3 Trigger words @zbrush_art Image Processing Parameters Repeat 20 Epoch 10 Save Every N Epochs 1 Training Parameters Seed - Clip Skip - Text Encoder learning rate 0.000001 Unet/DiT learning rate 0.0001 LR Scheduler cosine Optimizer AdamW Network Dim 64 Network Alpha 32 Gradient Accumulation Steps - Label Parameters Shuffle caption - Keep n tokens - Advanced Parameters Noise offset 0.03 Multires noise discount 0.35 Multires noise iterations 6 conv_dim - conv_alpha - Batch Size - Sample Image Settings Prompt @alice_3d zbrush art, Digital 3D rendering of a muscular, dark-haired anime warrioress girl with green pants and a white strapless top, mid-kick on a jagged rock. Sampler euler