Image-to-Video
Diffusers
lora
wan
explosion
special-effects
action
cinematic
high-impact
template:diffusion-lora
Instructions to use artificialguybr/Explosion-Redmond-WAN2-I2V-14B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use artificialguybr/Explosion-Redmond-WAN2-I2V-14B with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline from diffusers.utils import load_image, export_to_video # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Wan-AI/Wan2.2-I2V-A14B", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("artificialguybr/Explosion-Redmond-WAN2-I2V-14B") prompt = "ship explosion. explosion" input_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/guitar-man.png") image = pipe(image=input_image, prompt=prompt).frames[0] export_to_video(output, "output.mp4") - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
Small file size? (78mb) - Is this intended/low rank or?
#1
by CCP6 - opened
Hey,
Thanks a lot for the LoRA you've been posting. I've actually been hoping for an explosion/fire type LoRA as Wan never seems to do that well out of the box.
I noticed the file sizes for the LoRA .safetensors here are pretty small at only 78mb each. Is there are reason they're not in the 300mb range per most at a common rank of 32? If anything I'd think fire/explosions would need a higher quality training setting (and source media) due to the the motion/color changes.
Are the 78mb files uploaded the proper final version?
Thanks for any insight!