Instructions to use Lightricks/LTX-Video with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Lightricks/LTX-Video with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline from diffusers.utils import load_image, export_to_video # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Lightricks/LTX-Video", dtype=torch.bfloat16, device_map="cuda") pipe.to("cuda") prompt = "A man with short gray hair plays a red electric guitar." image = load_image( "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/guitar-man.png" ) output = pipe(image=image, prompt=prompt).frames[0] export_to_video(output, "output.mp4") - Notebooks
- Google Colab
- Kaggle
Longer video?
I was able to make a tiny video with ComfyUI, it is about a second long, maybe shorter.
What settings could I use to make longer videos? Even if it runs for a long time?
I was able to get a bit longer by increasing the number of frames. ( not fps, but actual frames ), but once i get very far it just produces 'noise'. I suspect much like current audio models out there its a fundamental limitation due to their training materials.
I also swapped out the save module to another so it would save mp4.. that webp format sucks.
@BigDeeper Thanks to performance and quality improvements in LTXV this week, it’s now feasible to generate high-quality 12-14 second clips in LTXS while maintaining very competitive waiting times.
@BigDeeper Thanks to performance and quality improvements in LTXV this week, it’s now feasible to generate high-quality 12-14 second clips in LTXS while maintaining very competitive waiting times.
When you say "it's feasible" how does one actually do it with ComfyUI? Or I have to run some python to take advantage of better performance and quality?
I've been using 10 or 12 fps and shrinking video size to get more seconds, but it is kind of a gamble. :)
Rwmove cloths
