People have been asking about a problem with Ollama in Lightning AI and being confused about needing to install Ollama and the LLM models each time they reopen the studio in lightning AI. I personally also wondered whats going on. But in this video I show how to solve the issue.

If you want a shortcut you can use the Lightning AI Studio template with the installation already done here:
https://lightning.ai/openintegrator/s...

First note the persistence guidance here:
https://lightning.ai/docs/overview/st...

Installation commands below just replace in - install.sh with angle bracket install.sh to download the file.

curl -fsSL https://ollama.com/install.sh - install.sh

OLLAMA_INSTALL_DIR="/teamspace/studios/this_studio/"

/teamspace/studios/this_studio/bin/ollama

chmod +x install.sh

./install.sh

vim ~/.zshrc

PATH=$PATH:/teamspace/studios/this_studio/bin


source ~/.zshrc

ollama serve

ollama run llama3.1