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

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.

Dataset

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

License

Apache-2.0. Note that running inference with this LoRA requires the Lightricks LTX-2.3 base model, whose own license terms still apply at inference time.

Acknowledgements

• Lightricks for releasing LTX-2.3
• oumad/LTX-2 for the trainer fork