---
license: mit
---
### Tangles on Stable Diffusion
This is the `<cora-tangle>` concept taught to Stable Diffusion via Textual Inversion. The concept is based on my daughter drawing random color fields with her hands or a paint brush and then me filling in the rest of the space with geometric doodles. 

You can load this concept into the [Stable Conceptualizer](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/stable_conceptualizer_inference.ipynb) notebook. 

Sample results below and more results [here](https://imgur.com/user/ritaleonova/posts.html).

Seed: Lush alien aquarium in the style `<cora-tangle>`
![<cora-tangle> nightime cityscape](https://i.imgur.com/ivMLbQk.jpeg)
Seed: Nighttime city scape in the style `<cora-tangle>`
![<cora-tangle> nightime cityscape](https://i.imgur.com/Q9XP2eb.png)
Seed: Hand holding paintbrush in the style `<cora-tangle>`
![<cora-tangle> nightime cityscape](https://i.imgur.com/7kHm7xm.png)
Seed: Fox in the style `<cora-tangle>`
![<cora-tangle> fox](https://i.imgur.com/Eu2WoAE.png)
Seed: Baby bird in the style `<cora-tangle>`
![<cora-tangle> bird](https://i.imgur.com/bza6jwp.png)

Here were the seed images for the style:
![<cora-tangle> 0](https://huggingface.co/sd-concepts-library/tangles/resolve/main/concept_images/1.jpeg)
![<cora-tangle> 1](https://huggingface.co/sd-concepts-library/tangles/resolve/main/concept_images/5.jpeg)
![<cora-tangle> 2](https://huggingface.co/sd-concepts-library/tangles/resolve/main/concept_images/0.jpeg)
![<cora-tangle> 3](https://huggingface.co/sd-concepts-library/tangles/resolve/main/concept_images/4.jpeg)
![<cora-tangle> 4](https://huggingface.co/sd-concepts-library/tangles/resolve/main/concept_images/2.jpeg)
![<cora-tangle> 5](https://huggingface.co/sd-concepts-library/tangles/resolve/main/concept_images/3.jpeg)
![<cora-tangle> 6](https://huggingface.co/sd-concepts-library/tangles/resolve/main/concept_images/6.jpeg)
![<cora-tangle> 7](https://huggingface.co/sd-concepts-library/tangles/resolve/main/concept_images/7.jpeg)

You can train your own concepts and load them into the concept libraries using [this notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/sd_textual_inversion_training.ipynb).