# Issues with loading models in 8bit in Colab

**URL:** <https://discuss.huggingface.co/t/issues-with-loading-models-in-8bit-in-colab/65716>\
**Category:** Beginners\
**Created:** [December 13, 2023, 7:23pm UTC](https://discuss.huggingface.co/t/issues-with-loading-models-in-8bit-in-colab/65716 "2023-12-13T19:23:42Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Diseno221](https://avatars.discourse-cdn.com/v4/letter/d/5fc32e/32.png) [@Diseno221](https://discuss.huggingface.co/u/Diseno221)\
**Post date:** [December 13, 2023, 7:23pm UTC](https://discuss.huggingface.co/t/issues-with-loading-models-in-8bit-in-colab/65716/1 "2023-12-13T19:23:42Z")

</div>

Hello. I am using Google Colab. I run this code  
!pip install sentencepiece -q  
!pip install accelerate -q  
!pip install bitsandbytes -q  
from transformers import AutoTokenizer, AutoModelForCausalLM  
tk = AutoTokenizer.from\_pretrained(“microsoft/Orca-2-13b”)  
model =AutoModelForCausalLM.from\_pretrained(“microsoft/Orca-2-13b”, load\_in\_8bit=True)

but I always get error  
ImportError: Using `load_in_8bit=True` requires Accelerate: `pip install accelerate` and the latest version of bitsandbytes `pip install -i https://test.pypi.org/simple/ bitsandbytes` or pip install bitsandbytes` .  
They are already installed. Why this happens?