How to use from the
Use from the
Diffusers library
pip install -U diffusers transformers accelerate
import torch
from diffusers import DiffusionPipeline
from diffusers.utils import export_to_video

# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("wan-ai/Wan2.1-T2V-14B-Diffusers", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("Muapi/kitakoji-hisui-wan2.1-14b-t2v")

prompt = "A man with short gray hair plays a red electric guitar."

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

Kitakoji Hisui/鍖楀皬璺儝銈广偆 | Wan2.1 14B T2V

preview

Base model: Wan Video 14B t2v Trained words: Kitakoji Hisui, anime-style girl. The image features an anime-style character with long, flowing teal hair that has a slight wave to it. The hair is styled in a way that it falls over the shoulders and back, with some strands framing the face. The character has striking orange eyes with black pupils, giving them a fierce and intense look.The character is wearing a black outfit with various straps and buckles, giving it a somewhat edgy and combat-ready appearance. The top part of the outfit includes a choker and a chain link design.The top has a strap across the chest, which is adorned with a small, gold-colored pendant shaped like a star. The strap also features a small, green gemstone. The character also wears black gloves and has a small, green ribbon tied around one wrist.

馃 Usage (Python)

馃攽 Get your MUAPI key from muapi.ai/access-keys

import requests, os
url = "https://api.muapi.ai/api/v1/wan21_t2v"
headers = {"Content-Type": "application/json", "x-api-key": os.getenv("MUAPIAPP_API_KEY")}
payload = {
    "prompt": "masterpiece, best quality",
    "lora_model": "kitakoji-hisui鍖楀皬璺儝銈广偆-wan21-14b-t2v",
    "lora_strength": 1.0,
    "width": 832,
    "height": 480,
    "num_frames": 81
}
print(requests.post(url, headers=headers, json=payload).json())
Downloads last month
8
Inference Providers NEW
This model isn't deployed by any Inference Provider. 馃檵 Ask for provider support