---
base_model:
- Wan-AI/Wan2.1-T2V-1.3B
- Wan-AI/Wan2.1-I2V-14B-480P
---
Wan video 14b 480p lora trained for i2v (but also supports t2v)
- trained for 14 hours using 3090
- trained in resolution: fp8 224x384 (33 and 49 frames) batch 1
  
Wan video 1.3b lora trained for t2v (but also supports i2v using vace)
- trained for 10 hours using 3090
- trained in resolution: bf16 432x768 (33, 49, 65 frames) batch 2

Dataset: 29 photos and 13 videos

prompt: `hand_skirt, woman, 25yo, front view, light hair, standing, holding a phone, man's hand is pulling her skirt up, revealing panties, front view, woman in a skirt and crop top, some people at background
front view, camera zooms in and moves down`