joyfox / Wan2.1-Fight-LoRA

valiantcat LoRA for Wan2.1 14B I2V 720p

Model File and Inference Workflow

📥 Download Links:

• wan2.1-fight.safetensors - LoRA Model File
• wan_img2video_lora_workflow.json - Wan I2V with LoRA Workflow for ComfyUI

Using with Diffusers

pip install git+https://github.com/huggingface/diffusers.git

import torch
from diffusers.utils import export_to_video, load_image
from diffusers import AutoencoderKLWan, WanImageToVideoPipeline
from transformers import CLIPVisionModel
import numpy as np

model_id = "Wan-AI/Wan2.1-I2V-14B-720P-Diffusers"
image_encoder = CLIPVisionModel.from_pretrained(model_id, subfolder="image_encoder", torch_dtype=torch.float32)
vae = AutoencoderKLWan.from_pretrained(model_id, subfolder="vae", torch_dtype=torch.float32)
pipe = WanImageToVideoPipeline.from_pretrained(model_id, vae=vae, image_encoder=image_encoder, torch_dtype=torch.bfloat16)
pipe.to("cuda")

pipe.load_lora_weights("valiantcat/Wan2.1-Fight-LoRA")

pipe.enable_model_cpu_offload() #for low-vram environments

prompt = "dajia,这是一个两个人对打得视频，镜头拉远，保持人物不变，这两个人开始对打，两个人不断旋转，拳打脚踢，姿势像是传统武术， 十分专业，两个人相互比试，打的有来有回，突然左边的人腾空跳起，保持飞在空中用腿不停踢另一个人上半身， 男人被踢得一边用手阻挡女人的飞踢一边后退，镜头跟随,保持背景和人物的服装不改变."

image = load_image("https://huggingface.co/valiantcat/Wan2.1-Fight-LoRA/blob/main/result/test.jpg")

max_area = 512 * 768
aspect_ratio = image.height / image.width
mod_value = pipe.vae_scale_factor_spatial * pipe.transformer.config.patch_size[1]
height = round(np.sqrt(max_area * aspect_ratio)) // mod_value * mod_value
width = round(np.sqrt(max_area / aspect_ratio)) // mod_value * mod_value
image = image.resize((width, height))

output = pipe(
    image=image,
    prompt=prompt,
    height=height,
    width=width,
    num_frames=81,
    guidance_scale=5.0,
    num_inference_steps=25
).frames[0]
export_to_video(output, "output.mp4", fps=16)