# Setting weights as adapter weights

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fsetting-weights-as-adapter-weights%2F82205%3E%5C
**Category:** 🤗Transformers\
**Created:** [April 17, 2024, 10:43am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fsetting-weights-as-adapter-weights%2F82205 "2024-04-17T10:43:27Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![AfonsoSousa](https://avatars.discourse-cdn.com/v4/letter/a/5f9b8f/32.png) [@AfonsoSousa](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FAfonsoSousa)%5C
**Post date:** [April 17, 2024, 10:43am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fsetting-weights-as-adapter-weights%2F82205%2F1 "2024-04-17T10:43:27Z")

</div>

I have a custom model with a T5 backbone and an additional GNN-based module. I would like to save and load just the module weights. As I understand, Transformers now has ways to handle this because of PEFT methods. How can I set my additional weights to be labeled as adapter weights for seamless saving and loading using the transformers methods (e.g., save\_pretrained() and PeftModel.from\_pretrained())?