Guides
The free route keeps your meeting recordings at home
A 7 minute meeting transcribed offline with the network blocked. Names and speakers are where it slipped.
7 minGuides
Guides
Free AI can answer your customer emails. Don’t leave it alone
Gemma 4 got 25 of 30 replies right in our shop test, and told one customer an address had changed when it had not.
7 min
Guides
Qwen3.6 on MLX returns an empty reply
Structured output from LM Studio came back blank five times out of five; the JSON sat in reasoning_content. Three ways round it.
8 min
Guides
Two agents, one Mac: MLX serves both at full speed
Two callers on Gemma 4 31B got 11.5 tokens a second each on MLX and 5.7 each on llama.cpp.
12 min
Guides
MuJoCo installs on a Mac in five seconds
Tested on an M4 Pro: pip install in 4.9 seconds, the fix for the viewer error, and 63,000 steps a second per core.
8 min
Guides
Kokoro narrates an article in a seventh of its length
Install Kokoro TTS on a Mac with no ffmpeg, a measured 7.1x real time on an M4 Pro, and a job that narrates new pieces on its own.
6 min
Guides
One skills folder can serve four coding agents
Where Claude Code, OpenCode, Hermes Agent and Pi each look, and the 106 dead paths that broke ours.
7 min
Guides
A picture costs a local model 22 times less
The same 64 by 64 board was 4,179 prompt tokens as hex and 187 as a PNG, with less than half the reasoning.
12 min
Guides
What a 48 GB Mac can do on ARC-AGI-3
The toolkit runs 4,000 actions a second; a 31B reasons at 11 tokens a second; one move costs 67 to 443 seconds.
16 min
Guides
A first ARC-AGI-3 agent in 40 lines
It survives resets, converts action ids, handles clicks and scores locally. Every guard was paid for by a failure first.
18 min
Guides
Why LM Studio returns an empty reply
Reasoning ate 4,093 of 4,096 tokens, four times in a row. Here is the fix.
22 min
Guides
What one ARC-AGI-3 move costs a local model
The MIT-licensed toolkit on a 48 GB M4 Pro with a local model, a single move measured, and who is winning the prize.
12 min
Guides
A MoE is five times faster and bigger in memory
A 26B mixture decodes 4 to 5 times faster than a dense model on a Mac and can take more RAM. Total parameters load, active ones compute.
10 min
Guides
Bonsai 27B or Gemma 4 31B on a Mac?
One is a fifth the size of the other, answers faster and holds more context; the hard questions decide it.
20 min
Guides
How to stop a background agent failing silently
Ours was dead for three days while cron kept firing. A heartbeat, a wrapper and a watchdog raise the alarm within the hour.
7 min
Guides
A local agent that briefs you every morning
Hermes Agent set up to fold calendar, goals and weather into one private daily briefing.
8 min
Guides
A private assistant that never leaves the room
Hermes Agent and LM Studio on an M4 Pro, with voice notes on Telegram at 3 am and nothing sent to a cloud.
17 min
Guides
Three agents, one brand voice
One voice spec wired into OpenCode, Hermes and Pi through AGENTS.md, so their output matches.
12 min
Guides
Small models drop tool calls; one did not
Local tool calling in Hermes and OpenCode on a 48 GB Mac mini: five models downloaded and deleted in a day, and the one that worked.
7 min
Guides
Gemma, Qwen and Llama speeds on an M4 Pro
Tokens per second, memory used, and the point where the machine starts to struggle.
9 min
Guides
How to run Gemma 4 12B with Ollama
The pull command, the memory it needs, the context you can afford, and how fast it really answers.
9 min
Guides
OpenClaw runs on a local model. Here is the setup
A private assistant on our test machine, and what it still cannot do.
9 min
Guides
OpenCode with a local model, offline
No internet, no API key: the setup, the config and the trade-offs we hit on a Mac.
13 min
Guides
LTX-Video on a Mac delivers, slowly
Hours with the local video model in ComfyUI on Apple Silicon: what consumer hardware gets and where it still falls short.
8 min
Guides
Marker turns PDFs into clean Markdown
Layout detection keeps tables and equations intact, on your own machine.
10 min
Guides
Piper speaks in under a second on any CPU
A lightweight text-to-speech engine built for a Raspberry Pi, with nothing leaving your machine.
9 min
Guides
Safe file access for a coding agent
An agent that edits files and runs commands is useful and, handled carelessly, risky. How we set its permissions.
10 min
Guides
A deep research engine that never leaves your machine
A model on Ollama plus a self-hosted search index that searches, reads and writes a cited report.
12 min
Guides
How much RAM does a local model need?
The commonest mistake is pulling a model your machine cannot hold. Sizing a model to the memory you have.
10 min
Guides
ComfyUI for beginners
It looks intimidating because it shows every step of image generation as a node. That is also where its power is.
11 min
Guides
Chat with your documents, locally
AnythingLLM: drag in a PDF, ask a question, get an answer with citations, on your own machine.
11 min
Guides
WhisperX tells you who said what
Whisper turns audio into text; WhisperX adds speakers and timestamps, offline.
11 min
Guides
Aider edits your code from the terminal
With a local model it reads the repo, proposes diffs, runs commands and commits with clean messages.
11 min
Guides
Copilot on your own laptop, with Continue.dev
Local autocomplete in VS Code through Ollama: you pick the model and no line of code leaves the room.
10 min
Guides
A private ChatGPT in ten minutes
Open WebUI wired to Ollama: a chat app that looks like ChatGPT, answered by a model on your machine.
10 min