Hero Run LoRA for Wan2.1 14B I2V 480p

Overview

This LoRA is trained on the Wan2.1 14B I2V 480p model.Follows the subject at high speed to capture intense, fast-paced motion. Ideal for chase scenes, athletic sequences, or adding urgency and kinetic energy to a shot.

Features

Trained on the Wan2.1 14B 480p I2V base model
Consistent results across different object types
Simple prompt structure that's easy to adapt

Community

Generate videos with 100+ Camera Control and VFX LoRAs on the Remade Canvas.
Discord: Join our community to generate videos with this LoRA for free

Prompt: A muscular athlete is poised in a sprinter's starting position on a red running track within an outdoor stadium under bright lights. He wears a blue and white athletic tank top and black shorts. Positioned low and facing the camera, he intensely looks forward. He then explodes from the starting position, beginning the 4ct3ion Action Run directly towards the camera. His body transitions from a crouch to a running stance, pumping his arms.

Prompt: A man wearing a tan shirt, vest, and pants with a bandana across his face runs directly toward the camera through a desert environment with fire and smoke. The camera follows the man in a fast-paced 4ct3ion Action Run, emphasizing his determined expression as he advances through the fiery environment.

Prompt: A young man wearing a blue t-shirt is sprinting directly towards the camera through a sunlit forest, his mouth open in a look of surprise or fear. He begins the 4ct3ion Action Run directly towards the camera, his arms pumping as he runs. The camera maintains a close, head-on perspective, with the background blurring slightly due to his motion.

Model File and Inference Workflow

📥 Download Links:

Hero_run.safetensors - LoRA Model File
wan_img2vid_lora_workflow.json - Wan I2V 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: 4ct3ion Action Run

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 I2V 480p
Training Data: Trained on 50 seconds of video comprised of 10 short clips (each clip captioned separately) of hero runs.
Epochs: 35

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!

Remade-AI
/

Hero-run