Local inference, kept alive on purpose
From a cold rig to a paid product, with nothing left to guess.
Athomic Smithy is one playbook and the scripts that go with it: how to run a local model 24/7 on a single Apple Silicon Mac, and sell something from it.
The chapters are the parts nobody else finishes — watchdog, tunnel, queueing, fallback, cost arithmetic. Every recipe names the rig, the software versions and its sources, and states plainly whether it has been run on our rig or only read out of the vendor documentation.
It does not promise a single dollar of passive income.
$47 Playbook·$197 Playbook + Bench·15 recipes
- Reference rig
- Mac mini M4, 16 GB
- Also covered
- M4 Pro, 24–48 GB
- Playbook
- $47
- Playbook + Bench
- $197
The model is not what stops you.
A developer looks at an idle M-series Mac with more unified memory than most cloud instances, and knows local models are now good enough. Then the questions start. The answers are scattered across walkthroughs, forum threads and vendor docs.
- 01Which quantized model actually fits 16 GB and still answers usefully.
- 02How to serve it behind a stable OpenAI-compatible endpoint.
- 03How to reach a home machine from the internet without opening the router.
- 04How to keep it alive through reboots, thermal throttling, macOS updates and sleep.
- 05How to queue requests when two users arrive in the same second.
- 06What to do when the machine is genuinely down and paying customers are waiting.
- 07What a thousand requests cost on your own rig, against the API bill.
Nothing joins those dots from a cold machine to a product with paying users. So the rig stays idle, or the stack gets assembled over three weekends and abandoned after the first 3 a.m. outage.
Four things to do with a Mac that has spare memory.
One of them is being sold loudly right now. It is the weakest of the four, and the numbers are public.
Rent the rig's idle compute to a token network
Reported by real providers on Apple Silicon networks, against calculators promising $75, at single-digit utilization.
Keep paying per token to a hosted API
The margin on everything you ship goes to the model vendor. None of it accrues to hardware you already own.
Rent cloud GPUs, or rent a hosted Mac mini
rentamac.io from $3.30 per day, myremotemac.com from $85 per month. Both reintroduce the monthly bill you were trying to kill.
Sell a product served from hardware you own
The marginal cost of a request is power. The playbook shows that arithmetic rather than asserting it.
Sourced from public provider reports and vendor pricing pages, not run on our rig. Figures as of 2026-08-30.
Every recipe ends with a line like this, or it does not publish.
The validation stamp names the rig, the memory, the operating system, the server version and the date it was last run. When something was read in a vendor document instead of run here, the recipe says exactly that.
Validated on Mac mini M4, 16 GB, macOS 15.3, Ollama 0.5.7 — 2026-04-02
Rig
The exact machine and its unified memory.
Versions
Verbatim and complete, never approximate.
Date
A revision, re-run when the software moves.
One spine, in the order you will need it.
Nine recipes carry a rig from a sealed box to something a stranger can pay for. Each one names the machine it was run on.
- 01Choose the modelBy task and memory budget, for each rig size.
- 02Serve the endpointA stable OpenAI-compatible surface on the rig.
- 03Expose the rig without opening a portA tunnel, never port forwarding.
- 04Keep the endpoint alive after a rebootlaunchd, a watchdog and a restart policy.
- 05Queue and cap concurrencyOne accelerator, many callers.
- 06See the rig get sick firstObservability before customers notice.
- 07Degrade instead of dyingA documented cloud fallback for outages.
- 08Do the unit-cost arithmeticCost per million tokens against current API pricing.
- 09Pick a product shape that worksHigh volume, low glamour, privacy-bound work.
Two objects. One price each, paid once.
Playbook
$47- Recipes 01–09 as a PDF, plus the scripts and config files.
- Lifetime access to every later revision.
- A ten-message taste of the Bench, when the Bench opens.
Paid once. Download opens immediately.
Playbook + Bench
$197- Everything in the Playbook, plus the extended set: recipes 10–15.
- Twelve months of the Bench, the assistant that answers from the Playbook.
- A fair-use ceiling of 60 messages per day, published so inference stays predictable.
Paid once. 15 recipes, and the Bench when it opens.
Checkout is hosted by NanoCorp, the merchant of record for Athomic Smithy. That is the name on the payment page and on your card statement. Downloads open on this site the moment the payment clears, and the link is emailed to you as well.
Not included
- No Intel Mac coverage.
- No Windows or Linux.
- No promise of revenue.
- No one-to-one support.
Both packs are open. Here is exactly what a stamp means today.
Every recipe carries a validation stamp. In revision 2026-08-30 all 15 are sourced: assembled from the vendor documentation named on the recipe, at the versions named, and not yet run on our rig. That is the stamp you will read on each one, in those words.
Validation stamps replace them as recipes are run end to end on the reference rig, and every revision is free to you for life. That is what the price buys: the assembled path now, and the measurements as they land.
If a recipe does not work on your rig, write within 30 days and the full amount comes back. Refunds are not conditional on how much you read.
Copying the address is all this button does. Nothing is charged, nothing is collected.
- Revision
- 2026-08-30
- Recipes written
- 15
- Stamped sourced
- 15
- Stamped validated
- 0
- Packs on sale
- 2 of 2