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
Load Latent Upscale Model and VAELoader KJ issue
Hi RuneXX, I am using LTX-2.3_-_I2V_T2V_Basic.json and have encountered two issues:
- The Load Latent Upscale Model can find ltx-2-spatial-upscaler-x2-1.0.safetensors in the directory, but cannot find ltx-2.3-spatial-upscaler-x2-1.0.safetensors, while the Load Upscale Model can find the two.
- I used ltx-2-spatial-upscaler-x2-1.0.safetensors to execute the workflow, but the VAELoader KJ node reported an error, whether it's Kijai's or unsloth's vae model:
Is there something wrong with me?
And the latent upscaler is here : https://huggingface.co/Lightricks/LTX-2.3/tree/main
Also a little "guide" here on where to find them and where to save them:
https://huggingface.co/RuneXX/LTX-2.3-Workflows/discussions/10
Have you updated ComfyUI to the latest?
First of all thank you runexx.
After upgrading ComfyUI from 0.15.1 to 0.16.4.
The first issue still exists. The files have been downloaded and placed in the specified directory. Can help check the result, the two loaders detect different results.


The second problem has been solved. It can run successfully whether using ComfyUI's or KJ's vae loader.
Now the issue is that the ltx2 upscaler is obviously not suitable. Although the process can run through, the effect is not good.
I found out that there is also a latent_upscale_models folder that should be put there, hahaha.
yes you are not the first that did his ;-).
But you figured it out, its a different upscaler



