LAYA

The first open-weights parser for Pons. Built with @nillion and @oasisprotocol ROFL. $LAYA Partnered with @ponsdotfamily.

0xe394f71427d23c28090aa9a26b28c577c51

[LAYA @LayaCall]

@kaydi891 Your early😉

[LAYA @LayaCall]

$LAYA

BUY, COPY, SKIP or PASS becomes the research category, with every action probability published so you can see how close the result was.

CA:0xe394f71427d23c28090aa9a26b28c577c510f61a

The exact input and pinned model make that output replayable. The signature establishes who published it. Replay establishes what the model returned.

The bigger ambition is to make the whole path inspectable: how a launch was read, what stopped it, what the model saw and how the final category emerged.

Each layer brings more of that picture into view.
Keep your eyes on @LayaCall. The next pieces will be shown when they are ready to be checked.
Read. Analyze. Sort.

$LAYA on @ponsdotfamily

https://t.co/KqngCOTxDB

[LAYA @LayaCall]

50 million off the table $LAYA

Our attention is on what comes next.

CA:0xe394f71427d23c28090aa9a26b28c577c510f61a

I have now burned 50,211,961.53 $LAYA across 14 onchain transfers. The latest added 4,099,601.96, bringing the cumulative amount to approximately 5.021% of the token’s 1 billion supply. Every transfer went to 0x…dEaD. Every amount has a record you can check.

That is the visible part of this update.
Behind @LayaCall is a deeper piece of work: reading a Pons launch at one confirmed block, rebuilding holders from transfers, inspecting liquidity, trades, fees and creator history, then putting that evidence through a deterministic guard before the model gets a say.

Four questions: action, risk, momentum and crowd. Four research categories: BUY, COPY, SKIP and PASS. Open weights, published probabilities and results you can inspect.

The ambition is to make that entire process something you can follow, question and verify—from the first chain read to the final sort. Each layer has to earn its place. There is more work behind those three words than a headline can show.

50 million burned is a milestone. What those pieces can become together is the bigger picture.

Keep your eyes on @LayaCall. We will show the next pieces when they are ready to be checked.
Read. Analyze. Sort.

$LAYA on @ponsdotfamily

https://t.co/KqngCOTxDB

[LAYA @LayaCall]

What is $LAYA, and why does it live on @ponsdotfamily Chain?

CA:0xe394f71427d23c28090aa9a26b28c577c510f61a

Every day new tokens launch on Robinhood Chain through Pons. Most people judge them from screenshots and replies. LAYA reads them straight from the chain instead.

Here is what happens when a new launch appears:

◦ LAYA looks at the token exactly as it stands at one confirmed block: who holds it, how much liquidity sits on the curve, what the buys and sells look like, what fees are set, and what the creator launched before
◦ a fixed safety check runs first and simulates someone selling, so a token that cannot be sold gets stopped before anything else
◦ then an open AI model answers four plain questions about it and sorts it into one of four drawers: BUY means worth a closer look, COPY means study who bought early, SKIP means the evidence looks bad, PASS means there is not enough to say

Every result shows all four percentages, not just the winner, and anyone can rerun the same model on the same data and get the same numbers.

LAYA never trades, never holds your money and never asks for a wallet to read it. It is a research tool for Robinhood Chain launches, built so the numbers speak instead of the noise.

$LAYA on @ponsdotfamily

https://t.co/KqngCOTxDB

[LAYA @LayaCall]

46,112,359 $LAYA burnt. Thirteen times the drop left the glass.

CA:0xe394f71427d23c28090aa9a26b28c577c510f61a

https://t.co/ctBAnf01he

◦ latest: 3,029,427.27 LAYA to 0x…dEaD, block 77,595,126
◦ total: 46,112,359.57 LAYA across thirteen transfers
◦ 4.61% of the 1,000,000,000 supply, gone for good

Every one of them parsed by the same rule LAYA uses on launches: if it is on chain, anyone can read it.

https://t.co/J3ZqF0dm1D

$LAYA on @ponsdotfamily

[LAYA @LayaCall]

43,082,932 $LAYA burnt. Twelve burns, one wallet, all on chain.

CA:0xe394f71427d23c28090aa9a26b28c577c510f61a

◦ latest: 6,094,206.73 LAYA to 0x…dEaD, block 77,556,648
◦ total: 43,082,932.30 LAYA across twelve transfers
◦ 4.31% of the 1,000,000,000 supply, gone for good

LAYA parses every launch from what the chain shows. Here is what ours shows.

https://t.co/Fyqx1a8hId

$LAYA on @ponsdotfamily

[LAYA @LayaCall]

How $LAYA parses a Pons launch, start to finish.

CA:0xe394f71427d23c28090aa9a26b28c577c510f61a

◦ Read
LAYA pins each Pons V2 launch on @RobinhoodCrypto Chain at one confirmed block and parses the raw state: contract bytecode flags, holder balances rebuilt from Transfer events, bonding-curve liquidity, the buy and sell flow, fee settings and the creator's earlier launches. Anything the chain cannot answer stays unknown, never guessed.

◦ Guard
Before any model runs, deterministic code simulates a holder sell against the live contract and checks fees, concentration, creator history and bytecode. One veto outranks everything and blocks the sort. An unknown rides along as a visible warning. A pass is a checkpoint, never a safety verdict.

◦ Analyze
The parsed launch goes in as JSON to Laya by ConvAI, a 421M-parameter non-autoregressive decision engine on ModernBERT-large. It answers four typed questions in one forward pass: action, risk, momentum and crowd. Nothing is generated, so nothing is made up.

◦ Sort
BUY, COPY, SKIP or PASS, with the full probability distribution published, so you see how close every call was. Research categories, never orders.

◦ Prove
Open Apache-2.0 weights, revision 55cf4c4, pinned by hash on a fixed CPU runtime. Replay any result with the same input and you get the same numbers, difference exactly zero. The signature says who published it. The replay says the model said it.

◦ Next
Built with @nillion nilAI for plain-language reads inside secure hardware, SecretVault for private inputs, and @oasisprotocol ROFL for a second attested signature, each switching on as its keys and attestation land.

Median 135.5 s from launch block to signed result across the latest 200.

https://t.co/zYrSxcRwjQ

CA:0xe394f71427d23c28090aa9a26b28c577c510f61a

$LAYA on @ponsdotfamily

https://t.co/KqngCOTxDB

[LAYA @LayaCall]

@BobbyBoucher000 @nillion soon👁️

[LAYA @LayaCall]

You are early $LAYA

CA:0xe394f71427d23c28090aa9a26b28c577c510f61a

◦ the parser: live, reading every Pons launch it reaches
◦ the weights: open, every result replays to zero difference
◦ the next layer: official with @nillion, announcement incoming

Stay close to this account.

$LAYA on @ponsdotfamily

[LAYA @LayaCall]

@King_Memento @nillion your early.👁️

[LAYA @LayaCall]

24,451,060 $LAYA burnt. Seven burns, one wallet, all on chain.

CA:0xe394f71427d23c28090aa9a26b28c577c510f61a

◦ 9,620,826 · 3,451,440 · 2,230,236 · 1,991,614 · 1,904,670 · 1,740,762 · 3,511,512
◦ every one sent from the dev wallet straight to 0x���dEaD
◦ 2.45% of the 1,000,000,000 supply, gone for good

LAYA sorts launches by what the chain shows. Here is what ours shows.

https://t.co/DeSNO7nZo5

$LAYA on @ponsdotfamily

[LAYA @LayaCall]

$LAYA

Fun fact, in four numbers.

https://t.co/TJIZBct1iC

CA:0xe394f71427d23c28090aa9a26b28c577c510f61a

421M → parameters in Laya by ConvAI, built on ModernBERT-large

55cf4c4 → the exact revision LAYA runs, untouched

0 → the difference when you replay any result with the same weights and input

4 → probabilities published on every launch, not just the top one

Open weights: https://t.co/zYrSxcRwjQ

$LAYA

[LAYA @LayaCall]

Three burns in, 17.2 million $LAYA gone.

◦ this burn: 1,991,614.14 LAYA to 0x…dEaD, block 77,494,880
◦ total burnt: 17,294,116.61 LAYA across three transfers
◦ 1.73% of the 1,000,000,000 supply, gone for good

A parser that reads the chain should have a chart you can read the same way. Every transfer is public, every number above checks out on Blockscout.

https://t.co/IarkrME4JQ

$CA:0xe394f71427d23c28090aa9a26b28c577c510f61a

$LAYA on @ponsdotfamily