# Replace Stable Diffusion class-conditional text with rows of attributes

**URL:** <https://discuss.huggingface.co/t/replace-stable-diffusion-class-conditional-text-with-rows-of-attributes/70771>\
**Category:** 🧨 Diffusers\
**Created:** [January 27, 2024, 12:19am UTC](https://discuss.huggingface.co/t/replace-stable-diffusion-class-conditional-text-with-rows-of-attributes/70771 "2024-01-27T00:19:22Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![tpremoli](https://avatars.discourse-cdn.com/v4/letter/t/3be4f8/32.png) [@tpremoli](https://discuss.huggingface.co/u/tpremoli)\
**Post date:** [January 27, 2024, 12:19am UTC](https://discuss.huggingface.co/t/replace-stable-diffusion-class-conditional-text-with-rows-of-attributes/70771/1 "2024-01-27T00:19:22Z")

</div>

I want to use Stable Diffusion model weights to generate class-conditional images- however, I don’t want these images to be conditional on a text prompt, but rather on a number of binary class attributes/rows.

In order to do this, I was thinking of using Diffusers, as it seemed the most straightforward. My thinking was to replace the CLIP text encoder/tokenizer with a custom encoder which maps the attribute rows into the latent space, however I can’t seem to find resources on this online, and was wondering if it was possible/feasible within the Diffusers library.

I understand that the StableDiffusionPipeline is likely too strict, however, I was wondering how I would define a model with these attribute rows as the conditioner for the generation, and how this model could be trained/fine-tuned.