<h2 id="motion-forge-wan-2.2-14b-t2i-(beginner-friendly)">MOTION FORGE WAN 2.2 14B T2I (Beginner-Friendly)</h2><h2 id="required-models-and-exact-locations">Required Models &amp; Exact Locations</h2>Main Checkpoint:<ul><li><code>Wan2.2-T2V-A14B-LowNoise-Q8_0.gguf</code> → <code>ComfyUI/models/unet/</code></li></ul>LoRA Files (all optional, can toggle on/off):<ul><li><code>lightx2v_I2V_14B_480p_cfg_step_distill_rank128_bf16.safetensors</code> → <code>ComfyUI/models/loras/wan_loras/</code> (Strength: 2.0)</li><li><code>dslr_video_00003.safetensors</code> → <code>ComfyUI/models/loras/wan_loras/</code> (Strength: 1.0)</li><li><code>Vintage-Cinematic-Redmond-WAN22.safetensors</code> → <code>ComfyUI/models/loras/wan_loras/</code> (Disabled by default)<ul><li>⚠️ Important: Create subfolder <code>wan_loras</code> inside loras directory</li></ul></li></ul>VAE:<ul><li><code>pig_wan_vae_fp32-f16.gguf</code> → <code>ComfyUI/models/vae/</code></li></ul>Text Encoder (CLIP):<ul><li><code>cow-umt5xxl-q8_0.gguf</code> → <code>ComfyUI/models/text_encoders/</code></li></ul>Upscale Model (optional):<ul><li><code>RealESRGAN_x2plus.pth</code> → <code>ComfyUI/models/upscale_models/</code></li></ul><h2 id="required-custom-nodes:">Required Custom Nodes:</h2><ol><li>ComfyUI-GGUF</li><li>rgthree-comfy (Power Lora Loader, Image Comparer)</li><li>ComfyUI-Easy-Use</li><li>ComfyUI_Memory_Cleanup</li><li>ComfyUI-KJNodes (WAN Video NAG)</li><li>ControlAltAI-Nodes (FluxResolutionNode)</li><li>ComfyUI WAN Video Nodes (core)</li></ol><h2 id="workflow-features:">Workflow Features:</h2><ul><li>Purpose: Generate high-quality images/videos from text prompts</li><li>Resolution: Flexible (768x768 default, multiple presets available)</li><li>Steps: 8 (optimized for speed)</li><li>VRAM: 10-14GB recommended</li><li>Beginner Friendly: Power Lora Loader allows easy LoRA toggling</li><li>Includes: 2x upscaling with before/after comparison</li></ul><hr /><h2 id="shared-models-between-workflows">Shared Models Between Workflows</h2>These models are used by both workflows (only need one copy):<ul><li><code>pig_wan_vae_fp32-f16.gguf</code> (VAE)</li><li><code>cow-umt5xxl-q8_0.gguf</code> (CLIP)</li><li><code>lightx2v_I2V_14B_480p_cfg_step_distill_rank128_bf16.safetensors</code> (LoRA)</li></ul><hr /><h2 id="complete-folder-structure">Complete Folder Structure</h2>ComfyUI/├── models/│ ├── unet/│ │ ├── HuMo-17b-Q8_0.gguf│ │ └── Wan2.2-T2V-A14B-LowNoise-Q8_0.gguf│ ├── loras/│ │ └── wan_loras/ ← CREATE THIS SUBFOLDER│ │ ├── lightx2v_I2V_14B_480p_cfg_step_distill_rank128_bf16.safetensors│ │ ├── dslr_video_00003.safetensors│ │ └── Vintage-Cinematic-Redmond-WAN22.safetensors│ ├── vae/│ │ └── pig_wan_vae_fp32-f16.gguf│ ├── text_encoders/│ │ └── cow-umt5xxl-q8_0.gguf│ ├── audio_encoders/│ │ └── whisper_large_v3_fp16.safetensors│ └── upscale_models/│ └── RealESRGAN_x2plus.pth

Files with SHA256: 1d65943b08aa6073a3f910e3a32e1fe8c09cfc666332b47ad48bebb13f942510