Stable Diffusion · LoRA

Atomic Explosion LoRA for WAN I2V 14B

LoRA for WAN I2V 14B that generates dramatic explosion effects from static images, with low and high noise variants for ComfyUI.

About this model

This LoRA specializes in creating dramatic, high-impact explosion effects from static images using the Wan-AI Wan2.2-I2V-A14B image-to-video model as its base. It transforms ordinary scenes into spectacular explosions with realistic physics, debris, fire, and smoke effects, making it ideal for action sequences and special effects work. The model is trained with two variants — low_noise and high_noise — to cover different noise schedules during video generation.

The LoRA is trained for image-to-video generation, meaning users provide a starting frame and the model animates an explosion sequence. The trigger phrase is EXPLOSION (as stated on the Hugging Face card) or ATOMIC EXPLOSION (as stated on the Civitai page). For best results, include the trigger phrase in the prompt describing the scene to be animated.

To use in ComfyUI, download both [WAN2.2-I2V]EXPLOSION_REDMOND_low_noise.safetensors and [WAN2.2-I2V]EXPLOSION_REDMOND_high_noise.safetensors (or their ATOMIC EXPLOSION variants from Civitai) and place them in ComfyUI/models/loras/. Add a Load LoRA node after the base model loader in the ComfyUI workflow, connect the model loader output to the LoRA node input, then connect the LoRA MODEL output to the KSampler. A recommended starting strength_model value is 0.8-1.0.

The model is released under the Creative Commons CC0 1.0 Universal license on Hugging Face, allowing for free commercial and non-commercial use. Special thanks for GPU time and sponsorship is credited to Redmond Ai.