An uncensored, hyper-realistic LoRA adapter for WAN 2.2 T2V 14B. Built for high-definition video generation with no content restrictions, aligned with Tetsuo AI's open-source ethos.

Uncensored and Open Source

This model is trained without safety filters or content restrictions. We believe generative AI should be open, transparent, and free from arbitrary censorship. Released under Apache 2.0 for personal and commercial use.

Hyper-Realistic Output

Fine-tuned with a focus on high-definition, photorealistic video rendering. WAN 2.2 supports up to 720p/1080p resolution with strong temporal consistency and motion quality. The model produces cinematic video output with realistic lighting, textures, and camera motion.

Character Training

The initial training run focuses on the Tetsuo AI character, a cyberpunk anime female rendered in two modes:

High-contrast B&W manga - Heavy ink work, aggressive linework, seinen manga aesthetic
3D cyberpunk portraiture - Volumetric lighting, neon blue highlights, rain, dark metallic surfaces

Trigger word: tetsuo_character

Trained using ComfyUI's built-in LoRA training nodes, the same unified pipeline as the Flux image model.

Usage

ComfyUI

Download the LoRA .safetensors file
Place it in ComfyUI/models/loras/
Load WAN 2.2 T2V model with the LoRA Loader node
Include tetsuo_character in your prompt

Example Prompts

tetsuo_character, cyberpunk woman walking through neon city, rain, volumetric lighting, cinematic

tetsuo_character, sitting at holographic terminal, typing, dark room, blue screen glow, focused expression

tetsuo_character, performing on stage, electric guitar, neon-lit venue, detached intensity, camera slowly zooming in

Training Details

Parameter	Value
Base Model	WAN 2.2 T2V 14B (Q8_0 GGUF)
Method	LoRA (Low-Rank Adaptation)
Rank	32
Steps	1000
Learning Rate	1e-4
Optimizer	AdamW
Trigger Word	`tetsuo_character`
Training Dtype	bfloat16
Bypass Mode	Yes (quantized base)
Framework	ComfyUI built-in training

License

Apache 2.0

piccassol/tetsuo-ai-wan-video-lora

Tetsuo AI - WAN 2.2 Video Generation LoRA