Skip to content

Instantly share code, notes, and snippets.

@msaelices
Last active December 2, 2023 21:44
Show Gist options
  • Select an option

  • Save msaelices/6c3eb284dbbab35aa55455da26bbdcbc to your computer and use it in GitHub Desktop.

Select an option

Save msaelices/6c3eb284dbbab35aa55455da26bbdcbc to your computer and use it in GitHub Desktop.
Installing Code LLama in Ubuntu
# We assume you have an Nvidia GPU
# Ubuntu dependencies
sudo apt install nvidia-cuda-toolkit
# From https://docs.anaconda.com/free/anaconda/install/linux/
apt-get install libgl1-mesa-glx libegl1-mesa libxrandr2 libxrandr2 libxss1 libxcursor1 libxcomposite1 libasound2 libxi6 libxtst6
# Install conda
# Download the installer from https://www.anaconda.com/download/#linux
# Execute it, entering “yes” to initialize Anaconda Distribution when asked
# Activate conda
source ~/anaconda3/bin/activate # change to the location you have decided to install conda
# Install the textgen conda environment
conda create -n textgen python=3.10.9
conda activate textgen
# Install pytorch
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu117
# Download Text generation web UI
git clone https://github.com/oobabooga/text-generation-webui
cd text-generation-webui
# Install dependencies
python -m pip install -r requirements.txt
# Checks you have CUDA enabled
python -m torch.utils.collect_env
# Run the web server
python server.py
# Navigate to http://127.0.0.1:7860/
# Click in the "model" tab and download this model: https://huggingface.co/TheBloke/WizardCoder-Python-13B-V1.0-GPTQ
# In the "parameters" tab, set the temperature to the minimum and the max tokens to somewhere around 2048
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment