SomBoy Style LoRA for Wan2.2-T2V

Style LoRA trained to transfer the SomBoy v2.5 art style (2.5D anime aesthetic from CivitAI) to Wan2.2-T2V.

Trigger Word

somboy_2d5

Usage

Use the trigger word somboy_2d5 in your prompt to activate the style.

Example prompts:

  • 1girl, portrait, long black hair, soft lighting, in the style of somboy_2d5
  • 1boy, casual outfit, modern aesthetic, in the style of somboy_2d5

Sample Videos

Sample 1: ๐Ÿ“น sample_01.mp4

Sample 2: ๐Ÿ“น sample_02.mp4

Sample 3: ๐Ÿ“น sample_03.mp4

Sample 4: ๐Ÿ“น sample_04.mp4

Sample 5: ๐Ÿ“น sample_05.mp4

Sample 6: ๐Ÿ“น sample_06.mp4

Sample 7: ๐Ÿ“น sample_07.mp4

Sample 8: ๐Ÿ“น sample_08.mp4

Sample 9: ๐Ÿ“น sample_09.mp4

Sample 10: ๐Ÿ“น sample_10.mp4

Training Details

  • Base Model: Wan2.2-T2V
  • Training Steps: 750
  • Learning Rate: 1e-4
  • Network Dim: 64
  • Dataset: 60 images generated with SomBoy v2.5 checkpoint

License

Apache 2.0

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support