Unlisted · noindex · optional education

Local Qwen on Bob’s Mini

LM Studio + Qwen 3.6 27B · optional · on-demand only · August 13, 2026

This page details the optional education card on Roxy’s status page. It is a teaching lab on Bob’s desk, not a second production brain and not tomorrow’s session work.

Status: optional plan only. Do not install it as part of proving cited answers or building company-event ingestion. Roxy on Nous Hermes Cloud stays the production assistant. LM Studio is the easy tool — not the fastest inference engine.

Why this exists

Bob has a 64 GB Mac Mini. A 27-billion-parameter local model fits it well and is a good way to see how on-device models behave — speed, context limits, when thinking helps, when it just burns tokens.

It is not a replacement for Cloud Roxy. The library, mail path, Honcho, and cited answers stay on Hermes Cloud. The Mini is a lamp you switch on, not a server that runs all day.

What to install

ItemChoiceWhy
AppLM Studio, Apple Silicon buildEasy UI, local OpenAI-compatible server, MLX runtime
Weights nowlmstudio-community/Qwen3.6-27B-MLX-8bit (~29.5 GB)Known-good Mac quant. Leaves headroom for macOS, chat, and context
Weights laterOfficial Qwen/Qwen3.8-27B, then LM Studio’s MLX 8-bit3.8 is not released. Community 3.8 repos today are placeholders — do not download them
Not firstBF16, 4-bit, or random GGUF forksBF16 is too tight on 64 GB; 4-bit wastes the RAM we have; GGUF is slower on this Mac

LM Studio is the on-ramp. A later, faster path on the same machine would be mlx_lm.server or rapid-mlx. That is not required for the education goal.

On-demand only

How it wires to Roxy

Load settings when Bob turns it on

What this is not