Hugging Face's logo

FuzzPuppy
/
LTX-2.3-Foley-LoRA

Diffusers
LTX.io
Safetensors
ltx-2.3
video-to-audio
foley
sound-effects
lora
Model card Files
xet
 Community
1
How to use from the
Use from the
LTX.io library
# 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 FuzzPuppy/LTX-2.3-Foley-LoRA --local-dir models/LTX-2.3-Foley-LoRA
hf download google/gemma-3-12b-it-qat-q4_0-unquantized --local-dir models/gemma-3-12b
# Text/image-to-video with the LoRA on the HQ two-stage base pipeline
uv run python -m ltx_pipelines.ti2vid_two_stages_hq \
    --checkpoint-path path/to/checkpoint.safetensors \
    --distilled-lora path/to/distilled_lora.safetensors 0.8 \
    --spatial-upsampler-path path/to/spatial_upsampler.safetensors \
    --gemma-root models/gemma-3-12b \
    --lora models/LTX-2.3-Foley-LoRA/<weights>.safetensors 1.0 \
    --prompt "your prompt here" \
    --output-path output.mp4
# For image-to-video, add: --image path/to/image.jpg 0 0.8

LTX-2.3 Foley

LTX-2.3 Foley is a video-to-audio LoRA for LTX-2.3. It adds realistic, visually synchronized Foley and sound effects to video without adding a music overlay.

This LoRA is especially useful when LTX-2.3 adds background music, score, or rhythmic soundtrack material but the desired output is audible real-world sound effects.

Tutorial

Watch the tutorial: using the LTX-2.3 Foley LoRA

Watch the tutorial on YouTube

Demo with and without LoRA

Compatibility

This LoRA is intended for use with LTX-2.3 base or distilled video-to-audio models.

License

This model is released under the LTX Community License configured on the Hugging Face repository.

Files

  • ltx-2.3-foley-400-steps.safetensors: 400-step Foley LoRA weights

Recommended Usage

Use a LoRA multiplier between 1 and 3.

Start around 1. If music or score still appears, increase the multiplier toward 2 or 3.

Prompt with a short description of the visible action in the video, followed by:

No speech is present. No music is present.

Example prompt:

A barista uses an espresso machine to steam milk. No speech is present. No music is present

Recommended negative prompt:

music, melody, song, singing, vocals, score, soundtrack, beat, rhythm bed, instrumental backing, tinny, thin, harsh, clipped, distorted, low bitrate

Examples

Example videos generated with the LoRA:

Door close

Pineapple slicing

Race car

Squash

Comparisons showing LTX-2.3 outputs with and without the Foley LoRA:

Barista comparison

Tennis comparison

ComfyUI Workflow

Use the companion ComfyUI workflow here: FuzzPuppy/LTX-2.3-Foley-Workflow

Downloads last month
4,416
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for FuzzPuppy/LTX-2.3-Foley-LoRA

Base model

Lightricks/LTX-2.3
Adapter
(93)
this model

Space using FuzzPuppy/LTX-2.3-Foley-LoRA 1