Instructions to use RuneXX/LTX-2.3-Workflows with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- LTX-2
How to use RuneXX/LTX-2.3-Workflows with LTX-2:
# Install the LTX-2 pipelines git clone https://github.com/Lightricks/LTX-2.git cd LTX-2 uv sync --extra natten
# Download weights from this repo # Substitute filenames from this repo's "Files and versions" if they differ hf download RuneXX/LTX-2.3-Workflows \ diffusion_models/<distilled-transformer>.safetensors \ text_encoders/gemma4-12b-with-proj-ltx-2.5-bf16.safetensors \ vae/<video-vae>.safetensors \ vae/<audio-vae>.safetensors \ latent_upscale_models/<spatial-upsampler>.safetensors \ latent_upscale_models/<temporal-upsampler>.safetensors \ --local-dir models/LTX-2.3-Workflows # DFR requires the detailing IC-LoRA (separate repo; strength is fixed at 0.5) hf download Lightricks/LTX-2.5-22b-IC-LoRA-Pixel-Spatial-Upscaler --local-dir models/LTX-2.5-22b-IC-LoRA-Pixel-Spatial-Upscaler# Distilled LTX-2.5 pipeline (fast) uv run python -m ltx_pipelines.distilled \ --transformer-path models/LTX-2.3-Workflows/diffusion_models/<distilled-transformer>.safetensors \ --text-encoder-path models/LTX-2.3-Workflows/text_encoders/gemma4-12b-with-proj-ltx-2.5-bf16.safetensors \ --video-vae-path models/LTX-2.3-Workflows/vae/<video-vae>.safetensors \ --audio-vae-path models/LTX-2.3-Workflows/vae/<audio-vae>.safetensors \ --spatial-upsampler-path models/LTX-2.3-Workflows/latent_upscale_models/<spatial-upsampler>.safetensors \ --num-frames 121 \ --prompt "A beautiful sunset over the ocean" \ --output-path output.mp4 # For image-to-video, add: --image path/to/image.jpg 0 0.8# DFR pipeline (higher detail fidelity; optional temporal 2x/4x) uv run python -m ltx_pipelines.dfr_pipeline \ --transformer-path models/LTX-2.3-Workflows/diffusion_models/<distilled-transformer>.safetensors \ --text-encoder-path models/LTX-2.3-Workflows/text_encoders/gemma4-12b-with-proj-ltx-2.5-bf16.safetensors \ --video-vae-path models/LTX-2.3-Workflows/vae/<video-vae>.safetensors \ --audio-vae-path models/LTX-2.3-Workflows/vae/<audio-vae>.safetensors \ --spatial-upsampler-path models/LTX-2.3-Workflows/latent_upscale_models/<spatial-upsampler>.safetensors \ --temporal-upsampler-path models/LTX-2.3-Workflows/latent_upscale_models/<temporal-upsampler>.safetensors \ --detailing-lora models/LTX-2.5-22b-IC-LoRA-Pixel-Spatial-Upscaler/ltx-2.5-22b-ic-lora-pixel-spatial-upscaler-x2-1.0.safetensors \ --spatial-upscalings 1 \ --temporal-upscalings 1 \ --height 1088 \ --width 1920 \ --num-frames 121 \ --prompt "A beautiful sunset over the ocean" \ --output-path output.mp4 # For 4K: --spatial-upscalings 2 --width 3840 --height 2176 # For image-to-video, add: --image path/to/image.jpg 0 0.8 - Notebooks
- Google Colab
- Kaggle
Workflow - I2V Multi-Subject Reference with Licon-MSR lora
** I2V Multi-Subject Reference with LTX2.3-Licon-MSR lora**
This is an early experimental workflow for using multiple reference inputs to create a video.
The references can be scene, subject, or objects. It helps to be very precise in the prompt and describe the inputs and what you want to happen
- Multiple characters from different reference images
- Character + object combinations
- Object + scene compositions
The lora is in test version and character consistency might improve in future updates of the lora.
Download the lora from here: https://huggingface.co/LiconStudio/LTX2.3-Mutiple-Subject-Reference
And feel free to give feedback to them about what works and what does not work (with the lora) https://huggingface.co/LiconStudio/LTX-2.3-Multiple-Subject-Reference/discussions
And this simple node (no new dependencies not already in comfy) https://github.com/liconstudio/ComfyUI-Licon-MSR
Feel free to try out the a workflow here:
https://huggingface.co/RuneXX/LTX-2.3-Workflows/tree/main
The workflow has 3 inputs, but you can increase up to 5 ref. inputs.
(since this is a brand new lora, improvements to wf might come as well, but should be fine for some testing already
Will try a variant with DEV model and more steps as well, might improve consistency
Sadly i always get a error like this: ValueError: guide pre_filter_counts (546) != keyframe grid mask length (819) at the CustomSamplerAdvanced node . Not sure what this means exactly...with the offical one it works but i needed to change the models and i like ur workflow more because of remove bg and all the cool stuff. And what does the frame_count mean in Licon MSR? If i set it do differend values i get differend numbers with the same errror, thanks in advance
ValueError: guide pre_filter_counts (546) != keyframe grid mask length (819)
strange error. Try update comfy, KJnodes etc.
But will reproduce my end. Never seen that error before, so maybe from the node from them..
You need this node https://github.com/liconstudio/ComfyUI-Licon-MSR but i guess you already have
Thanks u a right after updating kjnodes its working now :)
Hi.
I have tested this MSR lora/workflow and it's very good.
It would be interesting to mix it with prompt relay.
Do you have any workflow for that?
Thank you.