Instructions to use konakona/ltxvideo_q8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use konakona/ltxvideo_q8 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("konakona/ltxvideo_q8", 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
AssertionError: config must have _class_name=CausalVideoAutoencoder
#3
by hArtyom - opened
I did everything according to documentation, but getting this error:
AssertionError: config must have _class_name=CausalVideoAutoencoder
same here, I tried with original LTX vae ... but cannot make it work
I have fixed the problem. if you go to https://huggingface.co/Lightricks/LTX-Video/commit/c7c8ad4c2ddba847b94e8bfaefbd30bd8669fafc, you can see the config.json file has "CausalVideoAutoencoder" class. Copy this line, and replace the config.json in your vae folder with this line. This will resolve AssertionError: config must have _class_name=CausalVideoAutoencoder.