Run AI Models with Ollama in CI with GitHub Actions

Hi, I'm the founder of actuated and have been creating a bunch of materials on learning Firecracker for the community over the past couple of years. In this tutorial I wanted to show how you can use GPUs mounted via VFIO for CI - for AI use-cases including e2e testing and training/embedding with acceleration.

If you've not used ollama before, it also gives you a quick guide to try it out on a CPU or GPU locally.

We're looking for folks who could use GPU acceleration in GitHub Actions or GitLab CI - feel free to reach out to me via the form on the pricing page if that's of interest.

If not, hope you learned something.