Instructions to use ZuluVision/MoviiGen1.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use ZuluVision/MoviiGen1.1 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("ZuluVision/MoviiGen1.1", 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
Recommended Guidance and Shift Flow
#2
by DeepBeepMeep - opened
Many thanks for sharing your model with the community.
Do you have any recommended Guidance and Shift Flow values ? I would be grateful if you could share any tip, a few attempts I made have produced poor quality videos at 720p
For cinematic-level video generation, we recommend a resolution of 1920 * 832 (21:9) and a sample shift of 7 and default guidance. Since all our training process is held on a 4K-level dataset. And as for prompt writing, please refer to our tips in this repo.