# What is the best #1 caption model for adult industry? — retained record

> Retained reddit record from reddit:StableDiffusion.

Canonical: https://highsignal.app/data/records/a6cbcbe0eed35a65

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:StableDiffusion)  published 2026-10-18 20:10 ingested 2026-10-19 02:40

[original source ↗](https://www.reddit.com/r/StableDiffusion/comments/1x9d2r2/what_is_the_best_1_caption_model_for_adult/)

## retained text



```
What is the best image captioning model for dataset creation that truly understands concepts in the adult industry? I've tested Qwen3.6-VL and JoyCaption. Both seem decent, but neither is ideal. Some concepts are not recognized correctly. 

Now I'm wondering: Is there a model that performs really well for this use case, or would I get the best captioning results by fine-tuning a model on my own dataset? 

Thanks!
```



document fetched 2026-10-19 00:29 · canonical https://reddit.com/r/StableDiffusion/comments/1x9d2r2/what_is_the_best_1_caption_model_for_adult
