Instructions to use 2bidoubi/SPARK with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use 2bidoubi/SPARK with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("2bidoubi/SPARK", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
Simplify model card, add teaser, and update CVPR 2026 citation
Updates the approved model-card presentation: combines the full title into one heading, replaces the embedded video with the existing SPARK project teaser, removes “How SPARK works” and “Input and output,” and keeps setup concise with links to the paper and code. Replaces the arXiv BibTeX with the published CVPR 2026 citation (pages 7233–7243).
Content changes: README.md and media/spark-teaser.png. Hugging Face also added one .gitattributes entry for the teaser’s large-file storage. Model weights and license terms are unchanged. The previously uploaded video and poster are retained in the repository but no longer embedded.
The teaser uses a URL pinned to its upload commit so it displays both in this proposal and after merging. Its rendering was verified in the Hugging Face README preview.
refs/pr/2 ref