---
license: apache-2.0
---


# This is the example of [this PR](https://github.com/huggingface/diffusers/pull/4365)

Note that I trained only 2 epochs. I think more training epochs improve the quality of results.

# Dataset

I used dataset from [pokemon-blip-captions](https://huggingface.co/datasets/lambdalabs/pokemon-blip-captions)

# Example result

#### prompt = 'cute dragon creature'

Note that there is large influnce of random seed.

![image](image.png)