Join our Discord to train your own LoRAs and generate videos from them: https://discord.gg/7tsKMCbNFC
HUNYUAN v1.0 :
Trained on Hunyuan Video fp8 with 14 images of Boa Hancock in various episodes of One Piece.
The trigger words are: 80a_H4Nc0ck woman, Boa Hancock, black hair, One Piece
Recommended Settings:
LoRA strength = 1.0
Embedded guidance scale = 7.0
Flow shift = 17.0
Prompt Recommendations:
Best to keep the prompt relatively simple, here are some examples that have worked well for me:
Example 1: '80a_H4Nc0ck woman, Boa Hancock with black hair, One Piece, is waiting at a train station, her arms crossed as she leans against a pillar. A gust of wind rushes past as a train arrives, her long black hair flowing behind her.'
Example 2: '80a_H4Nc0ck woman, Boa Hancock with black hair, One Piece, is seated at a small wooden table in a quiet village café, stirring a cup of tea. She watches the steam rise with a calm expression, her golden earrings glinting in the morning light.'
Example 3: '80a_H4Nc0ck woman, Boa Hancock with black hair, One Piece, stands on the deck of the Kuja Pirates' ship, the salty ocean breeze lifting her long black hair. She slowly runs a hand through her hair, tilting her head back as she lets out a soft, amused laugh.'
If the video is realistic, i.e. not anime style, try slightly increasing the strength of the LoRA to 1.1-1.2 instead. If it doesn't quite look like Boa Hancock, try also describing more features that she has, like her golden earrings or her violet eyes.
Let me know if there are any questions, I'll be happy to help!