Video-to-Video
Diffusion Single File
LTX-2
English
clean-plate
object-removal
ltx-video
ic-lora
ltx-2.3
vfx
Instructions to use Lightricks/LTX-2.3-22b-IC-LoRA-Clean-Plate with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusion Single File
How to use Lightricks/LTX-2.3-22b-IC-LoRA-Clean-Plate with Diffusion Single File:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- LTX-2
How to use Lightricks/LTX-2.3-22b-IC-LoRA-Clean-Plate with LTX-2:
# 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 Lightricks/LTX-2.3-22b-IC-LoRA-Clean-Plate --local-dir models/LTX-2.3-22b-IC-LoRA-Clean-Plate hf download google/gemma-3-12b-it-qat-q4_0-unquantized --local-dir models/gemma-3-12b
# Video-to-video with the IC-LoRA (runs on the distilled base model) uv run python -m ltx_pipelines.ic_lora \ --distilled-checkpoint-path path/to/distilled_checkpoint.safetensors \ --spatial-upsampler-path path/to/spatial_upsampler.safetensors \ --gemma-root models/gemma-3-12b \ --lora models/LTX-2.3-22b-IC-LoRA-Clean-Plate/<weights>.safetensors 1.0 \ --video-conditioning reference.mp4 1.0 \ --prompt "your prompt here" \ --output-path output.mp4 - Notebooks
- Google Colab
- Kaggle
Degraded quality with LTX 2.5
#2
by HarryB - opened
I have noticed that running this lora on 2.5, produced reduced quality/fidelity on m y test.
I am using https://github.com/Lightricks/ComfyUI-LTXVideo/blob/master/example_workflows/2.5/LTX-2.5_V2V_ICLoRA_Single_Stage_Distilled.json
with theses model :
π ComfyUI/
βββ π models/
β βββ π text_encoders/
β β βββ gemma4-12b-with-proj-ltx-2.5-bf16.safetensors
β β βββ gemma4_e2b_it_bf16.safetensors
β βββ π loras/
β β βββ LTX-2.3-22b-IC-LoRA-Clean-Plate-1.0.safetensors
β βββ π diffusion_models/
β β βββ ltx-2.5-22b-distilled-transformer-bf16.safetensors
β βββ π vae/
β β βββ ltx-2.5-audio-vae-bf16.safetensors
β β βββ ltx-2.5-video-vae-conv-bf16.safetensors / ltx-2.5-video-vae-bf16.safetensors
Quality drops
Any ideas why this is happening ?
Unfortunately I post this video as I am uunder NDA
I have noticed a 2.5 version of the Lora and the quality is back. So I may consider this is compatibility issue ? May be it also the official workflow that differ between 2.3/2.5