Instructions to use RuneXX/LTX-2.3-2.5-Workflows with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LTX.io
How to use RuneXX/LTX-2.3-2.5-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-2.3-2.5-Workflows --local-dir models/LTX-2.3-2.5-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-2.3-2.5-Workflows/<distilled-checkpoint>.safetensors \ --spatial-upsampler-path models/LTX-2.3-2.5-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-2.3-2.5-Workflows/<checkpoint>.safetensors \ --distilled-lora models/LTX-2.3-2.5-Workflows/<distilled-lora>.safetensors 0.8 \ --spatial-upsampler-path models/LTX-2.3-2.5-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
Audio VAE invalid (just FYI)
I am using the I2V json. I have all the exact same models except those bae can't be found anywhere with those exact filenames. Your workflow both the video and audio vae have a 'KJ' in the name. The actual files on the kijai huggingface do not, I assumed it doesnt matter. Anyway everytime I run the graph I get "VAELoader
ERROR: VAE is invalid: None
If the VAE is from a checkpoint loader node your checkpoint does not contain a valid VAE." Highlighting the node loading the audio vae. When I select the older ltx2 audio vae, the graph runs but not with ltx23 audio vae. I saw you just updated that workflow so I redownloaded but this error persists.
So with the old audio vae it runs til the LTX2 NAG node which gives "LTX2_NAG
'function' object has no attribute 'to'" I dont know what it want, but since since it says optional I bypass it. After this no problems, generation time is fast, quality is nice, and even the old audio vae worked fine.
Yes I just named them KJ at end (Kijai's models), since i sometimes download more models to test if they work or not, and trying to organize it a little by source ;-)
But yes the models are from https://huggingface.co/Kijai/LTX2.3_comfy
If you get error with the LTX-2.3 audio vae you just need to update the KJNodes (this node pack also has the update NAG node)
https://github.com/kijai/ComfyUI-KJNodes/
(and update comfyUI as well, LTX-2.3 has changes that require update)
ah ok. My comfy is latest and greatest, I'll update kj nodes. Thank you. these workflows are very well structured
Edit: just want to confirm for anyone else seeing this, updating kj nodes fixed both issues