Instructions to use RuneXX/LTX-2.3-Workflows with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LTX.io
How to use RuneXX/LTX-2.3-Workflows with LTX.io:
# Install the LTX-2 pipelines git clone https://github.com/Lightricks/LTX-2.git cd LTX-2 uv sync --frozen
# Download the weights from this repo, plus the Gemma text encoder hf download RuneXX/LTX-2.3-Workflows --local-dir models/LTX-2.3-Workflows hf download google/gemma-3-12b-it-qat-q4_0-unquantized --local-dir models/gemma-3-12b
# Fast pipeline (distilled model, no distilled LoRA needed) uv run python -m ltx_pipelines.distilled \ --distilled-checkpoint-path models/LTX-2.3-Workflows/<distilled-checkpoint>.safetensors \ --spatial-upsampler-path models/LTX-2.3-Workflows/<spatial-upsampler>.safetensors \ --gemma-root models/gemma-3-12b \ --prompt "A beautiful sunset over the ocean" \ --output-path output.mp4 # For image-to-video, add: --image path/to/image.jpg 0 0.8# HQ pipeline (two-stage, higher quality) uv run python -m ltx_pipelines.ti2vid_two_stages_hq \ --checkpoint-path models/LTX-2.3-Workflows/<checkpoint>.safetensors \ --distilled-lora models/LTX-2.3-Workflows/<distilled-lora>.safetensors 0.8 \ --spatial-upsampler-path models/LTX-2.3-Workflows/<spatial-upsampler>.safetensors \ --gemma-root models/gemma-3-12b \ --prompt "A beautiful sunset over the ocean" \ --output-path output.mp4 # For image-to-video, add: --image path/to/image.jpg 0 0.8 - Notebooks
- Google Colab
- Kaggle
V2V Extend video with Reference Images
Hey, first of all thank you for your job, workflows are amazing. Wanted to ask, if there's a way to achieve v2v extend video with last frame or reference image to control where it goes? So far trying with prompts but it's messy sometimes.
Wan2GP somehow perfectly handled Continue video feature with end frame option, no brightness issues, freeze, no quality drops, it perfectly extends the video and ends at exact last frame, so curious if we can achieve it with any of your workflows, even without ending at exact last frame, just guiding it with references
The end frame is from the video input, or you mean set an end frame from an image you make?
I guess you mean a manually set last frame, so you guide where the extended video is going, and you provide the last frame via an image?
Thats totally doable, and might add a wf for that. Can be useful
To set the image i already have prepared, this way i'll have more ability to control how the video should end after extending
Yes that could be usable for sure. Will add a variant that has that ;-)
excellent, do you have it already or will take time to get it done?
excellent, do you have it already or will take time to get it done?
Had a few other things that was on the list, but should be able to add yours in not too long.
looking forward, thank you a lot
Will just tweak it a little bit, but from progress so far ;-)
Should hopefully work ok, can try it out here:
https://huggingface.co/RuneXX/LTX-2.3-Workflows/discussions/137