kabachuha/ltx2-disguise-drop

LTX-2 Disguise Drop!

Prompt

Disguise drop. The girl touches the top of her head with her hand and then she fully unzips her skin in two halves slowly pulling the zip with her hand across her body from the top of her head, splitting her face, to the waist to reveal a smug anime style anime girl. She throws away the previous unzipped skin.

Prompt

The video starts with multiple mice near a crushed cheese. Then suddenly the camera moves and focuses on the anthropomorphic mouse behind all them. The mouse stands on its two legs and, while the camera zooms out, the mouse touches the top of its gray head with its hand and then it fully unzips its skin in two halves slowly pulling the zip with its hand across its body from the top of its head, splitting its face, to the waist to reveal a fierce gigantic colossal ginger cat which roars furiously. The cat slowly increases in size as it is ejected from the mouse's gray skin. The previous unzipped skin falls down. The rest three of the mice run away in horror while the cat starts to chase them. Only one mouse unzips itself, the others keep the mouse form. There is only one skin layer. Dynamic pursuit scene. The cat is aggressive and angry and chases the mice.

Prompt

Disguise drop. A knight in shining, bulky armor stands in a castle courtyard. He struggles to move, then has an idea. He unzips the chest plate, revealing not a man, but a clever miniature raccoon inside. The miniature raccoon pops out, drops the suit, and scurries off with a piece of stolen treasure.

Prompt

The video starts with two anime girls: the girl on the left has yellow sweater, yellow hair, green eyes and purple trousers and the girl on the right has white shirt, black hair, blue eyes and blue skirt. The girl on the right says: "You are me!". Then the girl on the left decides to change her shape: the girl on the left touches the top of her own head with her hand and then she fully unzips her skin in two halves slowly pulling the zip with her hand across her body from the top of her head, splitting her face, to the waist to reveal a short girl looking just like the initial girl on the right: an anime girl with white shirt, black hair, sky blue eyes and a blue skirt. The previous unzipped skin falls down. The girl on the right looks at her curiously. The girl on the right says: "And I am you!". Now with the same looking two girls standing, a reciprocal situation happens: the girl on the right touches the top of her own head with her hand and then she fully unzips her skin in two halves slowly pulling the zip with her hand across her body from the top of her head, splitting her face, to the waist to reveal a tall girl looking just like the initial girl on the left: an anime girl with yellow sweater, yellow hair, green eyes and blue trousers. The previous unzipped skin falls down.

Prompt

Disguise drop. The girl touches the top of her head with her hand and then she fully unzips her skin in two halves slowly pulling the zip with her hand across her body from the top of her head, splitting her face, to the waist to reveal a realistic wolf. The wolf throws away the previous unzipped skin.

Prompt

Disguise drop. The anime girl touches the top of her head with her hand and then she fully unzips her skin in two halves slowly pulling the zip with her hand across her body from the top of her head, splitting her face, to the waist to reveal a realistic man. The man pulls off the skin, moving his legs out of it. He throws away the previous unzipped skin.

Model description

LTX-2 Disguise Drop!

Remake of my very first Wan lora now for LTX-2! The sound effects take it to a new level!

The LoRA brings the classic unzipping surprise cartoon effect where a character suddenly unzips their skin, revealing another character inside them all along!

This time - compared to my other LTX-2 loras so far - thanks to calibrated LR scheduling, it has virtually no grayness and you can use it in image2video and first-last-frame2video modes alike!

The workflows are embedded within the videos or you can use the .json file inside the Huggingface files folder. It's the native workflow with KJ Nodes and VideoHelperSuite.

The training recipe is exactly the same as for my previous one-hand-squish lora, except for the lowered rank to prevent overfitting. This lora converged in 3520 steps.

The model was trained with `Disguise drop.` as the prefix, but you can simply describe the action in details.

Trigger words

You should use Disguise drop. to trigger the video generation. Preferrably, describe the action according to the reference prompts, especially for complex scenes and situations.

Download model

Download them in the Files & versions tab.