Projekte.

Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.

3x faster speeds on MLX | Qwen 3.8 27B | Native MTP Speculative Decoding On Apple Silicon With No External Drafter.

Row-Bot - Personal AI Sovereignty. A local-first AI assistant with integrated tools, a personal knowledge graph, voice, vision, shell, browser automation, scheduled tasks, health tracking, and messaging channels. Run locally via Ollama or add opt-in cloud models. Your data stays on your machine.

Fulloch - The Fully Local Home Voice Assistant

A nightly newspaper that writes itself: an agent runs the desks, the figures come from code, and it refuses to publish an edition with marks in it.

A chat window for your own database. Users ask questions in plain English, Jeeves writes the SQL and runs it locally - the AI receives your schema structure only, never a row. Drop-in Blade widget, voice or text, any provider or a model you run yourself.