Working with new AI technology requires the ability to experiment and prototype, which requires repeatable deployments. This repo helps with the setup of llama, models and Open-webUI through the use of Ansible scripts.
- apt-update.yaml - Updates APT packages on an Ubuntu machine.
- cuda-install.yaml - Installs the CUDA Kit version 12.4.
- docker-install.yaml - Installs Docker directly from Docker.
- inventory.ini - An example inventory file that can be modified as needed.
- LICENSE - License file.
- llamacpp-install.yaml - Install playbook that downloads and compiles llamacpp [Warning: Work In Progress!]
- llamacpp-model-install.yaml - Playbook to install models desired for llamacpp.
- nginx-proxy-install.yaml - Installs the nginx proxy manager if desired
- ollama-install.yaml - Ollama install playbook that will provide a complete llama3 install with the latest llam3 8b model and Open-WebUI. Everything should be ready for use after it runs.
- README.md
TODOS:
- Creation of cloud VMs on Azure and Digital ocean via a playbook
- Update the video tutorial link once the YT video is complete
Link - https://tech.workfromhometech.io
Ubuntu Install for MAAS Setup: https://www.youtube.com/watch?v=ASgwQMN34JM
Link - https://youtu.be/7zls5w3Ar1U
