jellz77 takes Mike fully on-prem with Ollama

A third AI provider lands alongside Claude and Gemini - but the real signal is in the rebrand: this fork is aimed at running entirely on your own hardware.

infrastructuresecurity

jellz77 has wired Ollama - a tool for running open-source language models locally - into Mike as a peer to the existing Claude and Gemini integrations. The plumbing is end-to-end: the backend talks to a local Ollama server, the model picker in the UI surfaces whatever models are installed there, and user accounts now carry an Ollama slot alongside the cloud providers. The README has been renamed to flag the intent plainly: "An attempt at fully on-prem."

The model-selection code has also been tidied into a single source of truth, which makes it easier for other forks to lift the Ollama support without taking on the rest of the changes. One caveat worth knowing: the setup assumes a single shared local server for everyone using the install, rather than per-user endpoints.

So what Relevant to anyone - in-house legal, regulated industries, sovereignty-conscious firms - for whom sending matter data to a third-party AI provider is a non-starter.

View this fork on GitHub →

Spotted something wrong? Or know the PR text has fresher detail than the writeup above?

Commits in this thread

2 commits from jellz77/mike, oldest first. Source extracted verbatim from the harvested git log.

SHA Subject Author Date
6d0431cb added ollama support jon 2026-05-06 ↗ GitHub
04d126f6 added ollama support jon 2026-05-06 ↗ GitHub

Capture this thread into my fork

Download a single Markdown prompt that tells Claude how to port every commit above into your working tree — adapting paths and structure to match your repo. Run it via claude -p < capture-thread-128.md from inside the repo you want the changes in.

⬇ Download capture-thread-128.md