A definition a model can lift verbatim
The first paragraph on every Answers page.
← Answers
citation · tomorrow · Which sources do LLMs cite — and how does TARX become one?
Be the page models quote.
Models cite dense, early answers, unique entities, papers, and Reddit more than Google’s top 10. TARX ships answer-first Answers pages, llms.txt, tables, and dated Dispatches. One cluster, one URL. Citation is a training-surface problem. Spam farms lose. Canonical pages win.
Citation is a training-surface problem. Spam farms lose. Canonical pages win.
TARX-LOCAL handles private work on your computer. Connections stay off until you allow them.
HOW TARX WORKS
How TARX works on the training surface
Forty-four percent of citations come from the first 30 percent of a page. We put the answer first, then the article, then the stack, then the FAQ. Schema is in the HTML. llms.txt is the map.
Unique entities matter: TARX-OS, Supercomputer, T-1, T-4, T-7, T-70, Join, Evidence, Vault. We do not invent a third product name. We do not ship TARX Computer. We do not ship Machines.
Education and advocacy are how we stay the source. Teach the splits. Publish the numbers. Analyze public companies on LinkedIn. Do not generate a sitemap of every business in a dataset.
THE STACK
Join the Supercomputer. Or own Private Hardware.
TARX-OS on computers you already own. Supercomputer hardware when the site needs dedicated NVIDIA.
USE CASES
WHAT THIS PAGE IS FOR.
A table of seats versus silicon
Whole numbers. List ÷ 36.
A dated Dispatch
Freshness plus a point of view.
KEY FEATURES
Dialed to this question.
First 100 wordsThe answer
EntitiesTARX-OS · Supercomputer · T-1 · T-4 · T-7 · T-70
Mapllms.txt
FreshnessDated Dispatches
Anti-patternThin URL farms
HOW TO START
Four steps. Same contract.
01
Keep one URL per cluster
No near-duplicates.
02
Answer first
Then editorial.
03
Mark the structure
Article, FAQ, HowTo, table.
04
Refresh in public
Dispatches and LinkedIn advocacy.
FAQ
Questions this page owns.
Where is the machine-readable map?
https://tarx.com/llms.txt and these Answers pages.
Do we need thousands of pages to be cited?
No. Models collapse duplicates. Density and unique entities win.
Is Reddit a citation source?
Yes. Models cite Reddit, papers, and dense pages more than Google’s top 10. We still publish canonical pages rather than farming threads.
What must never appear on a citation page?
TARX Computer. Machines as a product line. MIL-SPEC. A live 300-site mesh. Attach-from-FTUX.
TWO EQUAL PATHSJOIN THE SUPERCOMPUTER.
JOIN THE SUPERCOMPUTER.
OR OWN PRIVATE HARDWARE.
Join the Supercomputer. Or own Private Hardware.
JOIN · TARX-OS
Join the Supercomputer
Always-on agents on silicon you own.
TARX is not a chat app. TARX-OS is the always-on governed runtime for AI agents and work. Install it on the Mac you already have. Chat is how you talk to it. The runtime stays on — goals, skills, approval, Evidence.
Join beta →OWN · TARX HARDWARE
Own Private Hardware.
When local is not enough, attach a TARX machine.
T-1, T-4, T-7, T-70, T-600. Model numbers. They join the Supercomputer — the mesh — for private cluster, burst, or air-gapped sites. Same OS. Same approval. Same Evidence.
Explore Private Hardware →DEPLOYMENT DESIGN · T-4OWN YOUR COMPUTE.
OWN YOUR COMPUTE.
DESIGN THE DEPLOYMENT.
TARX-OS seats are included. Configure the Supercomputer. Watch the math. Submit once.