vestademo.mp4
git clone https://github.com/Ednaordinary/Vesta/
cd Vesta
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
cd laion/DCXL
git-lfs https://huggingface.co/laion/CLIP-ViT-L-14-DataComp.XL-s13B-b90K
mv CLIP-ViT-L-14-DataComp.XL-s13B-b90K 0_CLIPModel
mv preprocessor_config.json ./0_CLIPModel
cd ../..
Then, create a ".env" with "DISCORD_TOKEN=" set to your token.
python3 ./main.py
The bot will start indexing all servers it is in straight away. Only embeddings are saved, so it shouldn't use too much storage. Your GPU will turn on a lot during this, as it downloads images as fast as possible and runs them in batches to the gpu. It only fetchs the messages and images it needs to, everything else is cached.