Remade-AI/Eye-Close-Up

Eye Close-Up LoRA for Wan2.1 14B T2V

Overview

This LoRA is trained on the Wan2.1 14B T2V model and allows you to generate videos depicting close ups of any kind of eye you describe! This actually also works well on the Wan2.1 I2V models. As long as a clear close-up of a face is presented, the video should zoom into and showcase the eye of the subject close-up, as in the second example video.

Features

• Trained on the Wan2.1 14B T2V base model
• Consistent results across different object and scenes types
• Simple prompt examples that are easy to adapt

Community

• Discord: Join our community to generate videos with this LoRA for free
• Request LoRAs: We're training and open-sourcing Wan2.1 LoRAs for free - join our Discord to make requests!

Prompt

3y3_cl053 eye extreme close up, The video shows a close up of a green cat-like eye with a shimmering emerald iris, resembling the Cheshire Cat.

Prompt

3y3_cl053 eye extreme close up, The video begins with a headshot of a woman with green eyes. The camera smoothly zooms in toward her face, focusing tightly on one eye. The camera is extremely close to the eye as she looks straight at the camera and blinks.

Prompt

3y3_cl053 eye extreme close up, The video shows a close up of a black eye with tiny white stars twinkling inside the iris, resembling a night sky.

Prompt

3y3_cl053 eye extreme close up, The video shows a close up of a bright red eye with smoke-like patterns swirling in the iris and a glowing orange pupil.

Model File and Inference Workflow

📥 Download Links:

• eye_10_epochs.safetensors - LoRA Model File
• wan_txt2vid_lora_workflow.json - Wan T2V with LoRA Workflow for ComfyUI

---

Recommended Settings

• LoRA Strength: 1.0
• Embedded Guidance Scale: 6.0
• Flow Shift: 5.0

Trigger Words

The key trigger phrase is: 3y3_cl053 eye extreme close up

Prompt Template

For prompting, check out the example prompts; this way of prompting seems to work very well.

ComfyUI Workflow

This LoRA works with a modified version of Kijai's Wan Video Wrapper workflow. The main modification is adding a Wan LoRA node connected to the base model.

See the Downloads section above for the modified workflow.

Model Information

The model weights are available in Safetensors format. See the Downloads section above.

Training Details

• Base Model: Wan2.1 14B T2V
• Training Data: Trained on 1 minute of video comprised of 18 short clips (each clip captioned separately) of various eye close-ups.
• Epochs: 10

Additional Information

Training was done using Diffusion Pipe for Training

Acknowledgments

Special thanks to Kijai for the ComfyUI Wan Video Wrapper and tdrussell for the training scripts!