# DORA memory requirements

**URL:** <https://discuss.huggingface.co/t/dora-memory-requirements/105805>\
**Category:** Beginners\
**Created:** [September 7, 2024, 7:57pm UTC](https://discuss.huggingface.co/t/dora-memory-requirements/105805 "2024-09-07T19:57:25Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![jaidyev799](https://avatars.discourse-cdn.com/v4/letter/j/f19dbf/32.png) [@jaidyev799](https://discuss.huggingface.co/u/jaidyev799)\
**Post date:** [September 7, 2024, 7:57pm UTC](https://discuss.huggingface.co/t/dora-memory-requirements/105805/1 "2024-09-07T19:57:25Z")

</div>

Hello Huggingface community! I am currently trying to use QDORA for finetuning Mistral-7B-Instruct. As per the documentation I have se the use\_dora flag to true -  
from peft import LoraConfig

config = LoraConfig(use\_dora=True, …)  
then why am I suddenly seeing an increase of 10GB during training when none of my training parameters have changed???