Remade-AI/Vintage-VHS

Vintage VHS Style LoRA for Wan2.1 14B T2V

Overview

This LoRA is trained on the Wan2.1 14B T2V model and allows you to generate videos in the style of vintage VHS!

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

vh5_c4m3r4 old VHS style, the video starts with a close-up shot of a wall covered in graffiti. The camera slowly pans to the right, revealing more of the wall and a door. It then pans left, revealing a man in a black t-shirt with his back to the camera. He reaches toward the door, then pulls it open. The video ends with a shot of the man walking through the doorway.

Prompt

vh5_c4m3r4 old VHS style, the video shows a sandy beach with large rocks and a blue sky with some clouds. The waves are crashing on the shore. A person walks across the sand. The video is distorted, with horizontal lines and colored dots around the edges of the screen.

Prompt

vh5_c4m3r4 old VHS style, the video starts with an overhead shot of a car driving down a straight, two-lane road with a white line down the middle. There is a wooden fence to the left of the road, followed by a field of dry grass. The grass is a yellowish-brown color. There are hills in the distance and a bright blue sky with white puffy clouds. The video is distorted, with horizontal lines and colored dots around the edges of the screen.

Model File and Inference Workflow

📥 Download Links:

• vhs_20_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: vh5_c4m3r4 old VHS style

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 3 minutes of video comprised of 45 short clips (each clip captioned separately) of various old clips.
• Epochs: 20

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!