AI Lenormand reading
A full Grand Tableau, drawn with true randomness and narrated by AI from the facts the engine computes — not from a prompt you type.
What makes it different
Most "AI tarot" tools ask a language model to invent a spread. Here the spread is not invented. The server deals all thirty-six cards into the traditional nine-by-four tableau, picks your significator, and calculates every relation that the reading is allowed to mention. The model's job is smaller and stricter: turn those verified facts and the curated knowledge base into prose you can actually read.
The practical difference: the draw is fixed before the model writes, so it cannot quietly swap in a card that suits a nicer story. It is instructed to describe only the computed positions, houses and relations, and never to add a card that is not on the table.
How a reading is built
1. The draw
Each reading uses a fresh 32-bit number from the operating system's cryptographic generator. That number deals the deck, and it is saved with the reading so the exact same tableau can be rebuilt for the PDF or when you reopen it later.
2. The geometry
A Grand Tableau reading is positional. The engine locates your significator, the house it occupies, the cards that touch it, the mirrored positions, the knight's moves, distances, the four corners, and the fate line. Those relations carry the meaning; the cards alone do not.
3. The knowledge base
Card meanings, houses, and all 1,260 directed card pairs come from a curated Lenormand knowledge base, with longer written interpretations for pairs that matter in your spread. The narration is grounded in those texts rather than generic model knowledge.
4. The narration
Only then does the model write. It is instructed to stay inside the supplied facts, address you in the second person, and respect the tone of each card. Failures never bill you: if the narrator is unavailable, the credit is returned automatically.
What you get
- The full nine-by-four tableau with card art and your significator highlighted
- Roughly 1,300–1,500 words covering the overall picture, past, present, what is coming, love, work, money and timing
- A PDF copy of any saved reading, including the tableau
- All of it saved in your account history
Your first full reading is free. After that, three readings cost $6 and ten cost $15; credits never expire.
Questions
Is this just ChatGPT writing a Lenormand reading?
No. The draw and the layout are deterministic code: the server deals all thirty-six cards into the nine-by-four Grand Tableau, selects the significator, and computes neighbours, houses, mirrors, knights, distances and the fate line. The AI only writes the narration from those verified facts and the curated card and pair knowledge base, and is instructed never to add, remove or move a card.
How are the cards drawn?
Each draw uses a fresh 32-bit value from the operating system's cryptographic random generator. There is no seeded shuffle, no question text affecting the deal, and no way to reproduce a previous draw except from the saved seed.
What is a Grand Tableau?
The full Lenormand spread: all thirty-six cards in four rows of nine. Instead of reading one card at a time, the Grand Tableau is read through position — which cards touch the significator, which houses they occupy, what mirrors and knight moves connect them, and how the fate line runs.
How long is the reading?
A full reading is roughly 1,300 to 1,500 words, organised into sections such as the significator and overall picture, the recent past, the present, what is coming, love, work and money, and timing.
Can I keep a reading?
Yes. Every reading is saved to your account and can be downloaded as a PDF that includes the tableau, the significator's house and the full text.
How much does it cost?
Your first full reading is free after signing in. After that, readings are sold in packs: three for $6 or ten for $15, in USD. Credits do not expire, and a reading is only charged when the narration succeeds.
Do you store my questions or use them to train AI?
Readings are stored so you can revisit them, and they are not used to train models. The draw does not depend on a question you type, so you can keep your question private.
Is an AI reading less accurate than a human reader?
The positions and relations are exact; the prose is written by a model from that verified context. Treat it as a structured, repeatable reading for reflection rather than a prediction, and remember it is for entertainment, not medical, legal or financial advice.