If you want a working WAN 2.2 workflow with loras (turbo and NSFW), this is really a simple one. Works perfectly with 16gb VRAM.should work on any comfyUI installation.use models/loras listed in workflow for SFW or download the ones named in workflow from civitAi for NSFW.You will need to download all models, vae, clip and loras yourself unfortunately.after that you only change input image, prompt, length (7s seems like optimal without problems with looping), height (i stay around 1024 for vertical and 720/800 for panoramic). rest doesn't seem worth changing.have fun.turbo (lightning) loras are here: <a target="_blank" rel="ugc" href="https://huggingface.co/lightx2v/Wan2.2-Distill-Loras/tree/main">https://huggingface.co/lightx2v/Wan2.2-Distill-Loras/tree/main</a>version might be different when You browse it.NSFW clip: <a target="_blank" rel="ugc" href="https://huggingface.co/Osrivers/nsfw_wan_umt5-xxl_fp8_scaled.safetensors/tree/main?not-for-all-audiences=true">https://huggingface.co/Osrivers/nsfw_wan_umt5-xxl_fp8_scaled.safetensors/tree/main?not-for-all-audiences=true</a>

Files with SHA256: 364b577bad6e25c6d960c8eab34af48dd35a0026f773c2c34ac0c7197c147b81