Image-to-Video
Diffusers
Safetensors
text-to-video
video-to-video
image-text-to-video
audio-to-video
text-to-audio
video-to-audio
audio-to-audio
text-to-audio-video
image-to-audio-video
image-text-to-audio-video
ltx-2
ltx-video
ltxv
lightricks
Instructions to use Lightricks/LTX-2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Lightricks/LTX-2 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-2", 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") - Inference
- Notebooks
- Google Colab
- Kaggle
LTX-2 β if this runs on mobile, it changes everything
#55
by 3morixd - opened
LTX-2 quality-to-compute ratio is the best for video generation. If this can be quantized for mobile, it would be the first practical on-device video gen model.
Our estimate: with 4-bit quant + frame skipping, LTX-2 could generate 3-second clips at ~15s compute on Snapdragon 865.
Lightricks, if you're reading this: we want to collaborate on mobile-optimized LTX.
- Dispatch AI (FZE), Sharjah UAE