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
distilled....screen distortion...unwanted spots appear in the video..How to fix?
I often get videos with this screen distortion...unwanted spots appear in the video I use "ltxv-13b-0.9.7-distilled-Q6_K.gguf" to create the video with 8 steps...this only happens with " distilled version"...but "distilled" is the best in time...the video is created in less than 2 minutes only while it takes about 6.5 minutes with 30 steps the non distilled version (with 8 gb GPU rtx 3060 ti) ...these video distortions happen in approximately 40% of the videos....Image to Video.. Video Example attached ⬇︎
The issue was fixed by using the official workflow... Base one ... and modifying the nodes required to load the VAE, model and clip (to load gguf) ......The good point is also that the video creation time has been reduced to 54 seconds ...8 steps...using the official Workflow....this one https://github.com/Lightricks/ComfyUI-LTXVideo/blob/master/example_workflows/13b-distilled/ltxv-13b-dist-i2v-base.json