macOS menu bar · local inference

Local models,
one click away.

Dakodeon lives in your menu bar. Start a local OpenAI-compatible router, point your agent at it, and switch models by id from the client — no runtime to bundle, no config to babysit.

$ brew install --cask emin93/tap/dakodeon
􀙇 100% Mon 9:41
Dakodeon Running
ACTIVE MODEL
unsloth/gemma-4-26B-A4B-it-qat-GGUF
Settings Logs Quit
Model manager

Download, serve, delete.

A built-in Settings window tracks every curated model. Watch downloads in real time and free up disk in a click — your agent switches models by id over the API, no picker to babysit.

Live download status

Progress, sizes, and hashes resolve straight from Hugging Face file metadata — no guessing, no stale numbers.

Selection lives in your agent

OpenCode (or any client) sends a model id and the router loads that profile with its configured assets and flags. Dakodeon just shows what's live.

Reclaim space

Delete weights you're done with. One confirm and the backing files leave the cache.

Settings
Models
unsloth/gemma-4-26B-A4B-it-qat-GGUFActive
15.69 GB · Vision · MTP draft
Delete
deepreinforce-ai/Ornith-1.0-35B-GGUF
21.17 GB
Downloading 24%5.08 GB / 21.17 GB
Cancel
Why Dakodeon

Small app. Real server.

OpenAI-compatible API

Serves llama-server at 127.0.0.1:8080/v1 with router presets for agents like OpenCode.

No bundled runtime

Uses the system llama.cpp and hf you already have. The app bundle stays tiny.

Curated profiles

Each profile ships ready to run with curated weights, optional companion assets, and tuned flags. No knobs to get wrong.

Get started

Install in one line.

$ brew install --cask emin93/tap/dakodeon

Requires llama.cpp and hf on your PATH · macOS 14+