README.md

---
license: mit
tags:
- stable diffusion
- text-to-image
---
This is a hypernetwork that generates pictures of food with the vintage recipe card look. Ingredients can be specified to give more specific results.

Dataset is images and text from https://linkin.bio/vintagerecipecards

#### vintagerecipecards_v1
* Trained with Stable Diffusion 2 model. (768-v-ema.safetensors)
* Hypernetwork learning rate = 0.00001
* 512x512 images

#### Example

<img src="https://i.imgur.com/ceDdW1M.png" width="300px">

```
<hypernet:vintagerecipecards-50000:1.2> dorito casserole, doritos, blue paint
Steps: 30, Sampler: DPM++ 2M Karras, CFG scale: 8, Seed: 3180085176, Size: 512x512, Model hash: 703d49a1d8, Model: 768-v-ema
```