How do you configure a Github Actions yml file to serve ollama and to load it with a base model (like llama3)? I tried to use a Dockerfile:

# Use an official Python runtime as a parent image
FROM python:3.11-slim

# Set the working directory in the container
WORKDIR /app

# Copy the current directory contents into the container at /app
COPY . /app

# Install Ollama
RUN curl -fsSL https://ollama.com/install.sh | sh

# Pull the Llama3 model
RUN nohup ollama serve > ollama.log 2>&1 & sleep 10 && ollama pull llama3

# Expose port 11434 to the outside world
EXPOSE 11434

# Start Ollama service and keep the container running
CMD nohup ollama serve > ollama.log 2>&1 & tail -f /dev/null

I tried to use it like this:

name: Ollama
 
on:
  pull_request:
    types: [opened, synchronize, reopened]
  workflow_dispatch:
 
jobs:
  auxiliary:
    runs-on: ubuntu-latest
    container:
      image: username/ollama_container:latest
      ports:
        - 11434:11434
    steps:
      - name: Checkout Repository
        uses: actions/checkout@v2

      - name: Check for dockerenv file
        run: (ls /.dockerenv && echo Found dockerenv) || (echo No dockerenv)

      - name: Configure Git Safe Directory
        run: git config --global --add safe.directory /__w/ai-integration/ai-integration

      - name: Start Ollama Service
        run: |
          nohup ollama serve > ollama.log 2>&1 &
          sleep 5
          curl -i http://localhost:11434 || (cat ollama.log && exit 1)
          tail -f /dev/null &

but if i don’t pull the llama3 model inside the workflow, it won’t be found, returning this error:

{"error":"model "llama3" not found, try pulling it first"}