Instructions to use Kijai/LTX2.3_comfy with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusion Single File
How to use Kijai/LTX2.3_comfy with Diffusion Single File:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Workflow: I2V & T2V - transfer body movements from reference video with IC-RealisDance lora
I2V & T2V - transfer body movements from reference video with IC-RealisDance lora
A nice alternative to the LTX Ic-Control lora many is probably familiar with.
And it uses a video input to drive the body movements, and a reference image to transfer the pose to.
Can often give more "natural" results, and less forcing the frame to change to adapt to the pose input.
LTX seems to work best with close up and not too fast motions, so might not do the most crazy tik-tok routines ;-)
(or you can try more steps, dev model, or even single pass if trying more "crazy" stuff)
In the workflow you can choose between LTX audio, custom audio input, or using the audio from the input video.
And the workflow has an easy toggle to choose between DW Pose and SD Pose.
For the lora, download the RealisDance here : https://huggingface.co/Kijai/LTX2-IC-LoRAs
( more about the concept here https://huggingface.co/theFoxofSky/RealisDance )
Feel free to play around with it ;-)
https://huggingface.co/RuneXX/LTX-2.3-Workflows/tree/main/Control-reference