---
author: ComfyTinker
baseModel: Hunyuan Video
hashes:
  AutoV1: E0397BFE
  AutoV2: 43CD08517B
  AutoV3: EA502BC7F222
  BLAKE3: 6ECBA4650ADEE945FD36D7262CDAE759AAFD89FA541B05931F1733AF4EBE734A
  CRC32: 679DFDF0
  SHA256: 43CD08517B7DB28FEE431900A7A647C42770AE456D353AD051BA66FD4690638D
metadata:
  format: SafeTensor
modelPage: https://civitai.com/models/1203396?modelVersionId=1368321
preview:
- https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/4839c378-01e9-42a2-8123-9161aac8e247/width=450/55651230.jpeg
website: Civitai
---

# Trigger Words

Spanks, Spanked, Spanking, Spank

# About this version

More flexible model that can be adapted to more novel scenarios and mixed with other LoRAs.

# Spanking - HunYuan

#Attention!

If you create or have a video that you think would be stellar to include in training, please let me know via DM or a link under the comments here. I'd love to continue retraining/refactoring this model using better and better videos. Anime videos don't show much jiggle (and therefor look weird), so good videos of those would be very helpful.

Trigger words:

```
Spanks, Spanked, Spanking, Spank
```

TIPS:

This model works best with "bent over" positions "lying" positions.  
  
LoRAs.

This version plays better with other LoRAs than the previous version, but it will need likely need more finetuning to accommodate other positions and styles that other LoRAs include.

CHANGES:

* Hands are more consistent
* Untrained objects can now be used to spank (like a tennis racket). Mileage may vary.
* Butts don't sway involuntarily
* Clothing can be changed more easily (Still heavily prefers exposed butt cheeks)
* More positions
* Larger training dataset (25 videos as opposed to 15)

Current known issues that:

* Skin is not captioned with a lot of detail, so it's not as dynamic as I'd like.
  
  + Need to add more captions that describe the skin in more detail. (Black, tan, white, smooth, red, tanlines, etc)
* Styles other than realistic are not represented.
  
  + Need to add captions to describe the style better (Realistic, well-lit, cartoon, etc)
* Using objects other than a hand are a bit harder to get working
  
  + Need to include detailed captions that describe the object being used to spank (hand, paddle, racket, belt, etc)
* Describing Multiple people in scene is currently difficult
  
  + Need to include more multi-person in dataset and describe each person in more detail.

Currently building a new LoRA training Workflow and will revisit this in a few weeks for retraining from scratch using a more varied dataset.