Vintage Cinematic (Redmond) LoRA para Wan

Instala las dependencias y carga el LoRA sobre el modelo base Wan2.2:

pip install -U diffusers transformers accelerate

import torch
from diffusers import DiffusionPipeline
from diffusers.utils import export_to_video

# Usa "mps" en dispositivos Apple compatibles
pipe = DiffusionPipeline.from_pretrained(
    "Wan-AI/Wan2.2-T2V-A14B",
    dtype=torch.bfloat16,
    device_map="cuda"
)
pipe.load_lora_weights(
    "artificialguybr/Vintage-Cinematic-Redmond-Wan22"
)

prompt = "Vintage Cinematic Color. A wide, static shot of a lone ballerina mid-spin in a vast, empty dance studio. Dramatic shafts of volumetric light illuminate rich brown tones and detailed wood textures, creating a melancholic arthouse film aesthetic."

output = pipe(prompt=prompt).frames[0]
export_to_video(output, "output.mp4")

En ComfyUI, coloca Vintage-Cinematic-Redmond-WAN22.safetensors en ComfyUI/models/loras/, añade un nodo Load LoRA después del cargador del modelo Wan y conecta su salida MODEL al KSampler. Un valor de strength_model entre 0.8 y 1.0 es un buen punto de partida.