Instructions to use lightx2v/Hy1.5-Distill-Models with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use lightx2v/Hy1.5-Distill-Models with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("lightx2v/Hy1.5-Distill-Models", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Diffusion Single File
How to use lightx2v/Hy1.5-Distill-Models with Diffusion Single File:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Link model to SGMD paper and improve model card documentation
#4
by nielsr HF Staff - opened
Hi! I'm Niels from the Hugging Face community science team.
This PR improves the documentation for the HunyuanVideo-1.5 distilled models by:
- Linking the models to the associated research paper: SGMD: Score Gradient Matching Distillation for Few-Step Video Diffusion Distillation.
- Fixing a typo in the metadata tags ("video genration" -> "video generation").
- Removing the redundant
pipeline_tagsmetadata field. - Ensuring the citation section includes the research paper details.
The distilled models enable efficient 4-step inference, and these changes help researchers and users find the underlying methodology.