SceneGen-Finetuned

This is a fine-tuned version of LTXV_2B_0.9.5 trained on custom data.

Model Details

• Base Model: LTXV_2B_0.9.5
• Training Type: LoRA fine-tuning
• Training Steps: 1500
• Learning Rate: 0.0002
• Batch Size: 1

Sample Outputs

Prompt A chef in a tall white hat is preparing elegant plated desserts with precision, while another man, holding a microphone, is narrating the scene like a TV show host.

Usage

This model is designed to be used with the LTXV (Lightricks Text-to-Video) pipeline.

🔌 Using Trained LoRAs in ComfyUI

In order to use the trained lora in comfy:

1. Copy your comfyui trained LoRA weights (comfyui..safetensors file) to the models/loras folder in your ComfyUI installation.
2. In your ComfyUI workflow:
   • Add the "LTXV LoRA Selector" node to choose your LoRA file
   • Connect it to the "LTXV LoRA Loader" node to apply the LoRA to your generation

You can find reference Text-to-Video (T2V) and Image-to-Video (I2V) workflows in the official LTXV ComfyUI repository.

Example Prompts

Example prompts used during validation:

• A chef in a tall white hat is preparing elegant plated desserts with precision, while another man, holding a microphone, is narrating the scene like a TV show host.

This model inherits the license of the base model (LTXV_2B_0.9.5).

Acknowledgments

• Base model by Lightricks
• Training infrastructure: LTX-Video-Trainer