Model Download Links (VPN required): https://huggingface.co/FX-FeiHou/wan2.2-Remix https://huggingface.co/NSFW-API/NSFW-Wan-UMT5-XXL/tree/main

Corresponding ComfyUI Workflow (Models not included) "myNSFW-Wan22Remix.zip" Link: https://pan.quark.cn/s/9e1c80b3c015 Extraction Code: ZRvy

Wan2.2_Remix_NSFW_i2v_14b_high_lighting_v2.0 is a customized version based on the open-source Wan2.2 video generation model developed by Alibaba's Tongyi Wanxiang team. It focuses on character dynamics, realistic motion, and scene consistency, supporting NSFW content generation.

Core Features

• Image-to-Video (I2V): Add motion effects (such as flowing hair or running rivers) to uploaded images. Supports NSFW themes without requiring additional LoRA settings.
• VRAM Optimization: Integrated based on the Work-Fisher workflow, supporting dual modes for both WebUI and ComfyUI. Runs with a minimum of 8GB VRAM.

Technical Highlights

• Enhanced Dynamics: The V2 version optimizes character actions, camera movements, and cinematic composition, enhancing motion smoothness and visual impact.
• Style Adaptation: Excels in styles such as Japanese anime and game CG, featuring vibrant colors and rich details.

Use Cases

• Anime & ACGN Content: Dynamic character design and fan-made video production.
• Art & Design: Concept art dynamic showcases and creative promotional short videos.