More upper body and full body and body anatomy zbrush arts and paintings. Model minicpmv4.5-en(natural language & English) Style Preference Custom Caption Prompt You are an expert image captioning assistant. For the given image, write one fluent English caption that describes only what is clearly visible: the main subject(s), key objects, setting, spatial relationships, colors/materials, lighting, style, and overall mood. Keep it factual, coherent, and about 120 tokens, never exceeding 150 tokens. Do not use tag lists, prompt commands, weights, or meta phrases (e.g., “this image shows”). Do not guess hidden details or read/transcribe text. Avoid camera/EXIF terms, file names, watermarks, and speculative words like “maybe” or “probably.” Output a single paragraph only. Parameter Settings Network Module LoRA Use Base Model REAL PHOTOGRAPHY (realism/photorealistic) - KREA 2 Trigger words @zbrush Image Processing Parameters Repeat 3 Epoch 8 Save Every N Epochs 1 Training Parameters Seed - Clip Skip - Text Encoder learning rate 1 Unet/DiT learning rate 1 LR Scheduler constant_with_warmup Optimizer Prodigy Network Dim 128 Network Alpha 64 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 @zbrush., A female figure with long red braided hair stands on a rocky surface, holding a large multi-barreled cannon. She wears a combination of leather and metal armor, including breastplates, arm guards, and greaves. The background is dark, with sparks and smoke emanating from the cannon and the ground around her. The figure has pointed ears and a determined expression. Sampler euler