Automation & AI agentsLocal AI hub
Local AI hub
An offline AI workshop on a single graphics card: 6 language models, speech recognition, two voice synthesisers, video upscaling, image recognition.
What it is
Almost everything in this section rests on one assumption: that there is a machine nearby able to think without a meter running. The local hub is that machine — a set of models arranged on shelves on a single graphics card, each for its own job, all available with no network and no account.
On the shelves: six language models of different sizes — from a fast one for small jobs to a heavy one for analysis; speech recognition; two different voice synthesisers; video upscaling; recognising what is in an image. Different sizes on purpose: keeping one big model for everything means paying its slowness for every trivial task.
Specifications: local models, speech recognition on whisper, voice synthesis on the spot, all on one graphics card, with no network. Assembled and checked on 21 Jul 2026.
Nearby work
Zero — an offline voice assistantA voice assistant that wakes on its name and thinks on the laptop’s graphics card: no internet, no login, an answer in 0.3–0.6 seconds.My own automation serverMy own server carrying over a hundred automations around the clock: bots, digests, finance, reminders, nightly backups, app synchronisation.Quality Watchdog — a guard that reads the automations themselvesA watchdog that reads how the automations are built rather than the run log, and catches what breaks silently.