Minimal CPU-only Ollama Docker Image
A 70 megabyte Docker image that runs Ollama on CPU only, built to be far smaller and faster to deploy than the official four-gigabyte Ollama image on systems without a GPU.
This project packages Ollama into a minimal Docker image built on wolfi.dev rather than Alpine, aimed at systems that only need CPU-based inference and do not have a GPU available. At about 70 megabytes, it is significantly smaller and faster to pull than the official Ollama image, while still exposing Ollama's normal API on the standard port once running.
It is meant to run anywhere a GPU is not available or not needed, such as local servers, edge devices, or cloud environments without GPU instances, for tasks like analyzing or streamlining logs with a local language model. Models are pulled and cached the same way as with the official image, and a separate web interface can be layered on top for chatting with the running model.