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
Add SPARK model card and 8-second preview video
Adds a concise model card with resource links, a method summary, input/output, checkpoint contents, setup instructions, licensing notes, and citation. Includes an 8-second silent excerpt from the existing SPARK project video and its poster.
Content changes are README.md, media/spark-preview.mp4, and media/spark-preview.jpg. Hugging Face also added one .gitattributes entry to track the video in large-file storage. Model weights, implementation code, and license files are untouched.
The embedded media links point to the uploaded media commit so playback works both in this proposal and after merging. The 8-second video was verified playing in Hugging Face’s README preview.
For maintainer review before merging: please confirm checkpoint provenance and the appropriate model-card license metadata. The existing README contains only a YAML license: mit field; the proposed README omits that field pending review and explains component-specific licensing in its License section. This is a README metadata change, not a change to the licensing terms of the code or checkpoints.
refs/pr/1 ref