ollama-docker

Why we included this project

This project packages Ollama into a Docker Compose setup that brings up the model server and a browser-based web UI together, so you can pull a model and start chatting within a few minutes. There's also a container full of LangChain examples for people who want to build on top of Ollama rather than just talk to it. GPU users get their own compose file and the NVIDIA Container Toolkit steps spelled out, while CPU-only machines run fine on the default config. Since everything stays local, it works as a quick way to test self-hosted inference or prototype an app, and the devcontainer means you can develop in the same environment. It's a thin wrapper around existing tools, not new software, but as a deployment blueprint it removes the fiddly parts.