Instructions to use aejion/AccVideo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use aejion/AccVideo with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("aejion/AccVideo", 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
Wan and beyond?
#5
by rp83 - opened
I am incredibly impressed with the speed and results from this model, it has improved generation time substantially, so thank you.
Once the HyVid I2V model is finished, are there plans to do this for Wan and/or others?
And, I think you may have a typo on the HuggingFace model card, the inference steps are listed as 50.
Nevermind, found the answer on GH, yes they are looking into Wan.
rp83 changed discussion status to closed