Instructions to use RuneXX/LTX-2-Workflows with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LTX.io
How to use RuneXX/LTX-2-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-Workflows --local-dir models/LTX-2-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-Workflows/<distilled-checkpoint>.safetensors \ --spatial-upsampler-path models/LTX-2-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-Workflows/<checkpoint>.safetensors \ --distilled-lora models/LTX-2-Workflows/<distilled-lora>.safetensors 0.8 \ --spatial-upsampler-path models/LTX-2-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
Checkerboard
I tried using some of your workflows and getting checkerboard in my videos.. any idea why?
Something with the model setup then
- If using DEV model (and not distilled model), you need to enable / add a DISTILL lora (if you want to keep using low steps and cfg)
- Using wrong vae
See here (and the workflows are made to work with these versions of the models):
https://huggingface.co/Kijai/LTXV2_comfy
and to add, if using distilled main model, you should bypass the distilled lora (ctrl + b)
Its a few things to trip on, since its so many models, but if all still fails, post a screenshot of your model loaders ;-)
alright thanks for quick response.. I will check it out.
this is what I see
your model loaders look all good. Thats odd.
One thing that comes to mind is perhaps you need to update both comfyUI, comfyui-gguf nodes and KJnodes
All 3 being important to have up to date, since the support for these models were added not that long ago
In ComfyUI manager find those nodes, and update. Or just click "update everything" in comfy manager
Odd, will make direct link to GGUF models that do work.
I did have a GGUF gemma model that didnt work here also, but that just produced an error, not checkerboard
But at least you figured it out ;-)
sorry, I may have given some bad info... I think maybe comfy had been running for almost a day, and it got into strange state,
I believe just restarting comfy is what actually fixed it.
I see now I can use the settings I sent you , and now it works fine..
No worries ;-) as long as all is working ;-)
torch bug , idk what triggers it was it few times as well , restart helps
