============================================================ INFERENCE START ============================================================ 🎲 Seed: 42 📦 Loading Transformer architecture... loaded 3D transformer's pretrained weights from VideoX-Fun/models/Diffusion_Transformer/Z-Image-Turbo/transformer ... ['VideoX-Fun/models/Diffusion_Transformer/Z-Image-Turbo/transformer/diffusion_pytorch_model-00001-of-00003.safetensors', 'VideoX-Fun/models/Diffusion_Transformer/Z-Image-Turbo/transformer/diffusion_pytorch_model-00002-of-00003.safetensors', 'VideoX-Fun/models/Diffusion_Transformer/Z-Image-Turbo/transformer/diffusion_pytorch_model-00003-of-00003.safetensors'] Missing keys will be initialized: ['control_all_x_embedder.2-1.bias', 'control_all_x_embedder.2-1.weight', 'control_layers.0.adaLN_modulation.0.bias', 'control_layers.0.adaLN_modulation.0.weight', 'control_layers.0.after_proj.bias', 'control_layers.0.after_proj.weight', 'control_layers.0.attention.norm_k.weight', 'control_layers.0.attention.norm_q.weight', 'control_layers.0.attention.to_k.weight', 'control_layers.0.attention.to_out.0.weight', 'control_layers.0.attention.to_q.weight', 'control_layers.0.attention.to_v.weight', 'control_layers.0.attention_norm1.weight', 'control_layers.0.attention_norm2.weight', 'control_layers.0.before_proj.bias', 'control_layers.0.before_proj.weight', 'control_layers.0.feed_forward.w1.weight', 'control_layers.0.feed_forward.w2.weight', 'control_layers.0.feed_forward.w3.weight', 'control_layers.0.ffn_norm1.weight', 'control_layers.0.ffn_norm2.weight', 'control_layers.1.adaLN_modulation.0.bias', 'control_layers.1.adaLN_modulation.0.weight', 'control_layers.1.after_proj.bias', 'control_layers.1.after_proj.weight', 'control_layers.1.attention.norm_k.weight', 'control_layers.1.attention.norm_q.weight', 'control_layers.1.attention.to_k.weight', 'control_layers.1.attention.to_out.0.weight', 'control_layers.1.attention.to_q.weight', 'control_layers.1.attention.to_v.weight', 'control_layers.1.attention_norm1.weight', 'control_layers.1.attention_norm2.weight', 'control_layers.1.feed_forward.w1.weight', 'control_layers.1.feed_forward.w2.weight', 'control_layers.1.feed_forward.w3.weight', 'control_layers.1.ffn_norm1.weight', 'control_layers.1.ffn_norm2.weight', 'control_layers.10.adaLN_modulation.0.bias', 'control_layers.10.adaLN_modulation.0.weight', 'control_layers.10.after_proj.bias', 'control_layers.10.after_proj.weight', 'control_layers.10.attention.norm_k.weight', 'control_layers.10.attention.norm_q.weight', 'control_layers.10.attention.to_k.weight', 'control_layers.10.attention.to_out.0.weight', 'control_layers.10.attention.to_q.weight', 'control_layers.10.attention.to_v.weight', 'control_layers.10.attention_norm1.weight', 'control_layers.10.attention_norm2.weight', 'control_layers.10.feed_forward.w1.weight', 'control_layers.10.feed_forward.w2.weight', 'control_layers.10.feed_forward.w3.weight', 'control_layers.10.ffn_norm1.weight', 'control_layers.10.ffn_norm2.weight', 'control_layers.11.adaLN_modulation.0.bias', 'control_layers.11.adaLN_modulation.0.weight', 'control_layers.11.after_proj.bias', 'control_layers.11.after_proj.weight', 'control_layers.11.attention.norm_k.weight', 'control_layers.11.attention.norm_q.weight', 'control_layers.11.attention.to_k.weight', 'control_layers.11.attention.to_out.0.weight', 'control_layers.11.attention.to_q.weight', 'control_layers.11.attention.to_v.weight', 'control_layers.11.attention_norm1.weight', 'control_layers.11.attention_norm2.weight', 'control_layers.11.feed_forward.w1.weight', 'control_layers.11.feed_forward.w2.weight', 'control_layers.11.feed_forward.w3.weight', 'control_layers.11.ffn_norm1.weight', 'control_layers.11.ffn_norm2.weight', 'control_layers.12.adaLN_modulation.0.bias', 'control_layers.12.adaLN_modulation.0.weight', 'control_layers.12.after_proj.bias', 'control_layers.12.after_proj.weight', 'control_layers.12.attention.norm_k.weight', 'control_layers.12.attention.norm_q.weight', 'control_layers.12.attention.to_k.weight', 'control_layers.12.attention.to_out.0.weight', 'control_layers.12.attention.to_q.weight', 'control_layers.12.attention.to_v.weight', 'control_layers.12.attention_norm1.weight', 'control_layers.12.attention_norm2.weight', 'control_layers.12.feed_forward.w1.weight', 'control_layers.12.feed_forward.w2.weight', 'control_layers.12.feed_forward.w3.weight', 'control_layers.12.ffn_norm1.weight', 'control_layers.12.ffn_norm2.weight', 'control_layers.13.adaLN_modulation.0.bias', 'control_layers.13.adaLN_modulation.0.weight', 'control_layers.13.after_proj.bias', 'control_layers.13.after_proj.weight', 'control_layers.13.attention.norm_k.weight', 'control_layers.13.attention.norm_q.weight', 'control_layers.13.attention.to_k.weight', 'control_layers.13.attention.to_out.0.weight', 'control_layers.13.attention.to_q.weight', 'control_layers.13.attention.to_v.weight', 'control_layers.13.attention_norm1.weight', 'control_layers.13.attention_norm2.weight', 'control_layers.13.feed_forward.w1.weight', 'control_layers.13.feed_forward.w2.weight', 'control_layers.13.feed_forward.w3.weight', 'control_layers.13.ffn_norm1.weight', 'control_layers.13.ffn_norm2.weight', 'control_layers.14.adaLN_modulation.0.bias', 'control_layers.14.adaLN_modulation.0.weight', 'control_layers.14.after_proj.bias', 'control_layers.14.after_proj.weight', 'control_layers.14.attention.norm_k.weight', 'control_layers.14.attention.norm_q.weight', 'control_layers.14.attention.to_k.weight', 'control_layers.14.attention.to_out.0.weight', 'control_layers.14.attention.to_q.weight', 'control_layers.14.attention.to_v.weight', 'control_layers.14.attention_norm1.weight', 'control_layers.14.attention_norm2.weight', 'control_layers.14.feed_forward.w1.weight', 'control_layers.14.feed_forward.w2.weight', 'control_layers.14.feed_forward.w3.weight', 'control_layers.14.ffn_norm1.weight', 'control_layers.14.ffn_norm2.weight', 'control_layers.2.adaLN_modulation.0.bias', 'control_layers.2.adaLN_modulation.0.weight', 'control_layers.2.after_proj.bias', 'control_layers.2.after_proj.weight', 'control_layers.2.attention.norm_k.weight', 'control_layers.2.attention.norm_q.weight', 'control_layers.2.attention.to_k.weight', 'control_layers.2.attention.to_out.0.weight', 'control_layers.2.attention.to_q.weight', 'control_layers.2.attention.to_v.weight', 'control_layers.2.attention_norm1.weight', 'control_layers.2.attention_norm2.weight', 'control_layers.2.feed_forward.w1.weight', 'control_layers.2.feed_forward.w2.weight', 'control_layers.2.feed_forward.w3.weight', 'control_layers.2.ffn_norm1.weight', 'control_layers.2.ffn_norm2.weight', 'control_layers.3.adaLN_modulation.0.bias', 'control_layers.3.adaLN_modulation.0.weight', 'control_layers.3.after_proj.bias', 'control_layers.3.after_proj.weight', 'control_layers.3.attention.norm_k.weight', 'control_layers.3.attention.norm_q.weight', 'control_layers.3.attention.to_k.weight', 'control_layers.3.attention.to_out.0.weight', 'control_layers.3.attention.to_q.weight', 'control_layers.3.attention.to_v.weight', 'control_layers.3.attention_norm1.weight', 'control_layers.3.attention_norm2.weight', 'control_layers.3.feed_forward.w1.weight', 'control_layers.3.feed_forward.w2.weight', 'control_layers.3.feed_forward.w3.weight', 'control_layers.3.ffn_norm1.weight', 'control_layers.3.ffn_norm2.weight', 'control_layers.4.adaLN_modulation.0.bias', 'control_layers.4.adaLN_modulation.0.weight', 'control_layers.4.after_proj.bias', 'control_layers.4.after_proj.weight', 'control_layers.4.attention.norm_k.weight', 'control_layers.4.attention.norm_q.weight', 'control_layers.4.attention.to_k.weight', 'control_layers.4.attention.to_out.0.weight', 'control_layers.4.attention.to_q.weight', 'control_layers.4.attention.to_v.weight', 'control_layers.4.attention_norm1.weight', 'control_layers.4.attention_norm2.weight', 'control_layers.4.feed_forward.w1.weight', 'control_layers.4.feed_forward.w2.weight', 'control_layers.4.feed_forward.w3.weight', 'control_layers.4.ffn_norm1.weight', 'control_layers.4.ffn_norm2.weight', 'control_layers.5.adaLN_modulation.0.bias', 'control_layers.5.adaLN_modulation.0.weight', 'control_layers.5.after_proj.bias', 'control_layers.5.after_proj.weight', 'control_layers.5.attention.norm_k.weight', 'control_layers.5.attention.norm_q.weight', 'control_layers.5.attention.to_k.weight', 'control_layers.5.attention.to_out.0.weight', 'control_layers.5.attention.to_q.weight', 'control_layers.5.attention.to_v.weight', 'control_layers.5.attention_norm1.weight', 'control_layers.5.attention_norm2.weight', 'control_layers.5.feed_forward.w1.weight', 'control_layers.5.feed_forward.w2.weight', 'control_layers.5.feed_forward.w3.weight', 'control_layers.5.ffn_norm1.weight', 'control_layers.5.ffn_norm2.weight', 'control_layers.6.adaLN_modulation.0.bias', 'control_layers.6.adaLN_modulation.0.weight', 'control_layers.6.after_proj.bias', 'control_layers.6.after_proj.weight', 'control_layers.6.attention.norm_k.weight', 'control_layers.6.attention.norm_q.weight', 'control_layers.6.attention.to_k.weight', 'control_layers.6.attention.to_out.0.weight', 'control_layers.6.attention.to_q.weight', 'control_layers.6.attention.to_v.weight', 'control_layers.6.attention_norm1.weight', 'control_layers.6.attention_norm2.weight', 'control_layers.6.feed_forward.w1.weight', 'control_layers.6.feed_forward.w2.weight', 'control_layers.6.feed_forward.w3.weight', 'control_layers.6.ffn_norm1.weight', 'control_layers.6.ffn_norm2.weight', 'control_layers.7.adaLN_modulation.0.bias', 'control_layers.7.adaLN_modulation.0.weight', 'control_layers.7.after_proj.bias', 'control_layers.7.after_proj.weight', 'control_layers.7.attention.norm_k.weight', 'control_layers.7.attention.norm_q.weight', 'control_layers.7.attention.to_k.weight', 'control_layers.7.attention.to_out.0.weight', 'control_layers.7.attention.to_q.weight', 'control_layers.7.attention.to_v.weight', 'control_layers.7.attention_norm1.weight', 'control_layers.7.attention_norm2.weight', 'control_layers.7.feed_forward.w1.weight', 'control_layers.7.feed_forward.w2.weight', 'control_layers.7.feed_forward.w3.weight', 'control_layers.7.ffn_norm1.weight', 'control_layers.7.ffn_norm2.weight', 'control_layers.8.adaLN_modulation.0.bias', 'control_layers.8.adaLN_modulation.0.weight', 'control_layers.8.after_proj.bias', 'control_layers.8.after_proj.weight', 'control_layers.8.attention.norm_k.weight', 'control_layers.8.attention.norm_q.weight', 'control_layers.8.attention.to_k.weight', 'control_layers.8.attention.to_out.0.weight', 'control_layers.8.attention.to_q.weight', 'control_layers.8.attention.to_v.weight', 'control_layers.8.attention_norm1.weight', 'control_layers.8.attention_norm2.weight', 'control_layers.8.feed_forward.w1.weight', 'control_layers.8.feed_forward.w2.weight', 'control_layers.8.feed_forward.w3.weight', 'control_layers.8.ffn_norm1.weight', 'control_layers.8.ffn_norm2.weight', 'control_layers.9.adaLN_modulation.0.bias', 'control_layers.9.adaLN_modulation.0.weight', 'control_layers.9.after_proj.bias', 'control_layers.9.after_proj.weight', 'control_layers.9.attention.norm_k.weight', 'control_layers.9.attention.norm_q.weight', 'control_layers.9.attention.to_k.weight', 'control_layers.9.attention.to_out.0.weight', 'control_layers.9.attention.to_q.weight', 'control_layers.9.attention.to_v.weight', 'control_layers.9.attention_norm1.weight', 'control_layers.9.attention_norm2.weight', 'control_layers.9.feed_forward.w1.weight', 'control_layers.9.feed_forward.w2.weight', 'control_layers.9.feed_forward.w3.weight', 'control_layers.9.ffn_norm1.weight', 'control_layers.9.ffn_norm2.weight', 'control_noise_refiner.0.adaLN_modulation.0.bias', 'control_noise_refiner.0.adaLN_modulation.0.weight', 'control_noise_refiner.0.after_proj.bias', 'control_noise_refiner.0.after_proj.weight', 'control_noise_refiner.0.attention.norm_k.weight', 'control_noise_refiner.0.attention.norm_q.weight', 'control_noise_refiner.0.attention.to_k.weight', 'control_noise_refiner.0.attention.to_out.0.weight', 'control_noise_refiner.0.attention.to_q.weight', 'control_noise_refiner.0.attention.to_v.weight', 'control_noise_refiner.0.attention_norm1.weight', 'control_noise_refiner.0.attention_norm2.weight', 'control_noise_refiner.0.before_proj.bias', 'control_noise_refiner.0.before_proj.weight', 'control_noise_refiner.0.feed_forward.w1.weight', 'control_noise_refiner.0.feed_forward.w2.weight', 'control_noise_refiner.0.feed_forward.w3.weight', 'control_noise_refiner.0.ffn_norm1.weight', 'control_noise_refiner.0.ffn_norm2.weight', 'control_noise_refiner.1.adaLN_modulation.0.bias', 'control_noise_refiner.1.adaLN_modulation.0.weight', 'control_noise_refiner.1.after_proj.bias', 'control_noise_refiner.1.after_proj.weight', 'control_noise_refiner.1.attention.norm_k.weight', 'control_noise_refiner.1.attention.norm_q.weight', 'control_noise_refiner.1.attention.to_k.weight', 'control_noise_refiner.1.attention.to_out.0.weight', 'control_noise_refiner.1.attention.to_q.weight', 'control_noise_refiner.1.attention.to_v.weight', 'control_noise_refiner.1.attention_norm1.weight', 'control_noise_refiner.1.attention_norm2.weight', 'control_noise_refiner.1.feed_forward.w1.weight', 'control_noise_refiner.1.feed_forward.w2.weight', 'control_noise_refiner.1.feed_forward.w3.weight', 'control_noise_refiner.1.ffn_norm1.weight', 'control_noise_refiner.1.ffn_norm2.weight'] Initializing missing parameter 'control_layers.10.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.8.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.11.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.8.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.5.after_proj.bias' with zero. Initializing missing parameter 'control_layers.9.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.5.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.after_proj.weight' with zero. Initializing missing parameter 'control_noise_refiner.1.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.after_proj.bias' with zero. Initializing missing parameter 'control_layers.0.before_proj.bias' with zero. Initializing missing parameter 'control_noise_refiner.0.before_proj.weight' with zero. Initializing missing parameter 'control_layers.8.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.9.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_layers.9.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_layers.4.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.11.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.after_proj.bias' with zero. Initializing missing parameter 'control_noise_refiner.0.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.after_proj.bias' with zero. Initializing missing parameter 'control_layers.14.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.14.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.5.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.5.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.14.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_layers.14.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.9.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.5.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.after_proj.bias' with zero. Initializing missing parameter 'control_layers.14.after_proj.bias' with zero. Initializing missing parameter 'control_layers.4.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.8.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_layers.3.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_layers.10.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.9.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.after_proj.weight' with zero. Initializing missing parameter 'control_layers.5.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.14.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.14.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.9.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.after_proj.weight' with zero. Initializing missing parameter 'control_layers.3.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.8.after_proj.weight' with zero. Initializing missing parameter 'control_layers.13.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.9.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.after_proj.weight' with zero. Initializing missing parameter 'control_layers.3.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.11.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.14.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.8.after_proj.bias' with zero. Initializing missing parameter 'control_noise_refiner.1.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.5.after_proj.weight' with zero. Initializing missing parameter 'control_layers.11.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.14.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.5.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.after_proj.bias' with zero. Initializing missing parameter 'control_layers.6.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.8.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.5.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_layers.8.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.after_proj.bias' with zero. Initializing missing parameter 'control_layers.12.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.8.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.9.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.9.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.9.after_proj.weight' with zero. Initializing missing parameter 'control_layers.5.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_layers.8.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.after_proj.bias' with zero. Initializing missing parameter 'control_noise_refiner.0.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.11.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_layers.11.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.11.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.after_proj.bias' with zero. Initializing missing parameter 'control_layers.6.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.14.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.after_proj.bias' with zero. Initializing missing parameter 'control_layers.8.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.11.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.8.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.14.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.11.after_proj.weight' with zero. Initializing missing parameter 'control_layers.5.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.14.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.11.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.after_proj.bias' with zero. Initializing missing parameter 'control_layers.6.after_proj.weight' with zero. Initializing missing parameter 'control_layers.11.after_proj.bias' with zero. Initializing missing parameter 'control_layers.14.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.after_proj.bias' with zero. Initializing missing parameter 'control_layers.11.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.9.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.14.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.after_proj.weight' with zero. Initializing missing parameter 'control_layers.7.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_layers.1.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.9.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.9.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.before_proj.weight' with zero. Initializing missing parameter 'control_layers.4.after_proj.weight' with zero. Initializing missing parameter 'control_noise_refiner.1.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.5.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.9.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.5.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.11.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.before_proj.bias' with zero. Initializing missing parameter 'control_layers.11.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.5.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.after_proj.weight' with zero. Initializing missing parameter 'control_layers.5.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.14.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.8.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.after_proj.bias' with zero. Initializing missing parameter 'control_layers.11.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.after_proj.weight' with zero. Initializing missing parameter 'control_layers.0.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.9.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.8.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.14.after_proj.weight' with zero. Initializing missing parameter 'control_layers.2.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.5.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_all_x_embedder.2-1.bias' with model.state_dict(). Initializing missing parameter 'control_layers.9.after_proj.bias' with zero. Initializing missing parameter 'control_layers.14.attention.to_v.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.after_proj.weight' with zero. Initializing missing parameter 'control_layers.5.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.feed_forward.w2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.11.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.feed_forward.w3.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.0.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_layers.12.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.2.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.attention.norm_q.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.11.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.6.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.after_proj.weight' with zero. Initializing missing parameter 'control_layers.13.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.8.adaLN_modulation.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_noise_refiner.1.feed_forward.w1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.3.attention_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.ffn_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.8.attention.to_q.weight' with model.state_dict(). Initializing missing parameter 'control_layers.13.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_layers.9.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.7.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.1.after_proj.weight' with zero. Initializing missing parameter 'control_layers.8.attention_norm1.weight' with model.state_dict(). Initializing missing parameter 'control_layers.4.attention.to_out.0.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.0.attention.norm_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_layers.2.adaLN_modulation.0.bias' with model.state_dict(). Initializing missing parameter 'control_layers.7.ffn_norm2.weight' with model.state_dict(). Initializing missing parameter 'control_layers.10.attention.to_k.weight' with model.state_dict(). Initializing missing parameter 'control_layers.12.attention.to_q.weight' with model.state_dict(). ### All Parameters: 9511.126848 M ### attn1 Parameters: 0.0 M 📦 Loading ControlNet Union weights... Loading from: controlnet_union_2.0.safetensors ✅ ControlNet Union loaded: 295 parameters Missing keys: 521, Unexpected keys: 0 📦 Loading LoRA lora_9178637007800530956.safetensors (scale=1.0) ✅ LoRA applied: 360 parameters updated 📦 Loading VAE & Encoders... Loading checkpoint shards: 0%| | 0/3 [00:00