---
license: cc-by-nc-4.0
library_name: diffusers
pipeline_tag: image-to-video
base_model: Lightricks/LTX-2.3
tags:
- ltx-2
- ltx-2.3
- text-to-video
- image-to-video
- lora
- googly-eyes
language:
- en
---

# Googly Eyes LoRA for LTX-2.3

Slaps oversized googly-eye stickers onto whatever subject is in your LTX-2.3 generation. Pure, dumb fun.

**Trigger word:** `googlyeyes` — that's all you need. The LoRA does its thing on the trigger word alone, you don't have to write any other prompt. You can add a regular prompt on top if you want to try and direct the scene somehow, no promises you'll succeed though; anyway the trigger just needs to be in there somewhere.

## Sample outputs

<table>
<tr>
<td>
<video controls width="100%">
  <source src="https://huggingface.co/TheBurgstall/ltx-2.3-googlyeyes-lora/resolve/main/samples/pulp.mp4" type="video/mp4">
</video>
</td>
<td>
<video controls width="100%">
  <source src="https://huggingface.co/TheBurgstall/ltx-2.3-googlyeyes-lora/resolve/main/samples/term.mp4" type="video/mp4">
</video>
</td>
</tr>
</table>

## Usage in ComfyUI

Standard LTX-2.3 IC-LoRA workflow. Drop the `.safetensors` into your ComfyUI `models/loras/` folder, add a **Load LoRA** node, set strength to 1.0, and put `googlyeyes` somewhere in your prompt.

Reference workflows are in the official [LTX-2 repository](https://github.com/Lightricks/LTX-2).

## Dataset

I built the entire dataset myself — fully synthetic, and curated and filtered by hand. No real people in the training data.

## License

CC BY-NC 4.0

## Acknowledgements

- [Lightricks](https://huggingface.co/Lightricks) for releasing LTX-2.3
- [`oumad/LTX-2`](https://github.com/oumad/LTX-2/tree/feature/prodigy-optimizer) for the trainer fork