01 · OWN
Mint a Neur0 Share
1,000 ERC-721 shares. Each one is 1/1,000 of every ETH and $NEUR0 the model ever earns, claimable at any time.
Think collectively. Train boldly.
Neur0 is one deep time-series Transformer trained on Robinhood Chain — fractioned into 1,000 shares, fed by contributors paid in $NEUR0, and sold as an API whose revenue flows straight back to the shares. The network is the asset.
01 · OWN
1,000 ERC-721 shares. Each one is 1/1,000 of every ETH and $NEUR0 the model ever earns, claimable at any time.
02 · FEED
Every epoch we measure how much your contribution improved validation loss. Rewards are posted as a Merkle root and paid in $NEUR0.
03 · TRAIN
A pre-norm Transformer encoder over 120-minute windows of on-chain candles, predicting 5, 15 and 60-minute returns. Each version's weight hash is published on-chain.
04 · EARN
Traders and bots buy API credits in ETH or $NEUR0. 70% goes to the RevenueVault for share holders, 30% to the treasury that funds the next epoch.

Trained on Uniswap V3 swap logs from Robinhood Chain: Pons launches, Stock Tokens, 24/7. Inputs are 1-minute candles with volume, trade count and buy ratio. Outputs are multi-horizon return and direction forecasts, signed by the API so anyone can verify which weights produced them.

Rendered by the NeuroShares contract itself from the token id. Seven palettes, three LED shapes, three flows, 3 to 7 activation centres, a rare signal row.

Three ways in. Each is scored by leave-one-out validation: we retrain without your contribution and measure what was lost. The bigger the loss, the bigger your share of the epoch budget.
Historical swaps, order flow, off-chain signals on Robinhood Chain assets.
Tag rugs, insiders, bots and real graduations. Consensus-checked.
Run a training node on your own hardware during retrain windows.
$NEUR0 is created on Pons V2, straight into a locked Uniswap V3 pool on Robinhood Chain. Fixed supply of one billion. It is what contributors earn, what API buyers can pay with, and what share holders receive when they do.


Every large model today is owned by the company that trained it. The data came from everyone; the value went to one place. Neur0 flips the direction: the network belongs to a thousand people, the ones who feed it get paid, and the weights that make money are the weights that are published.
Grown on Robinhood Chain. Rooted in its holders.