Instructions to use lightx2v/Wan2.1-T2V-14B-StepDistill-CfgDistill-Lightx2v with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use lightx2v/Wan2.1-T2V-14B-StepDistill-CfgDistill-Lightx2v 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("lightx2v/Wan2.1-T2V-14B-StepDistill-CfgDistill-Lightx2v", dtype=torch.bfloat16, device_map="cuda") pipe.to("cuda") prompt = "A man with short gray hair plays a red electric guitar." image = load_image( "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/guitar-man.png" ) output = pipe(image=image, prompt=prompt).frames[0] export_to_video(output, "output.mp4") - Notebooks
- Google Colab
- Kaggle
Noisy Images
#3
by Griphen116 - opened
Trying to use this with native ComfyUI, and I'm seeing only videos of noise.
Using the same T2V workflow with the rank32 version of the lora, and it works fine at 4 steps.
Anything changed that would cause issues with workflow compatibility?
Nevermind, I see the issue.
ComfyUI logs show that the lora blocks are not being loaded.
Same issue as the other post, i'll close this and wait for an answer in the other thread.
Griphen116 changed discussion status to closed