NFT marketplace & credit

NFT lending

Lend against an asset with no reliable price, and the interface is the least of your problems.

Open the live prototype ↗ demo data throughout

The why, in one breath

People own digital art worth real money but can’t borrow against it, because nobody can agree what it’s worth. This concept sizes the loan by what the art actually sells for — and plans the sale before the loan exists.

Premise

An NFT holder is asset-rich and liquidity-poor — crypto’s version of the borrower whose portfolio doesn’t count. Borrowing against the asset unlocks capital without a sale.

Why it’s hard

No reliable price — the “floor” is one thin order from fiction. Liquidity vanishes exactly when liquidations cluster. And the keenest borrowers know their valuation is generous.

The insight

When you can’t price the collateral, price the exit: what will it fetch, how fast, in a bad week? Card rails call this thin-file underwriting.

The underwriting design

A lending book that admits what it doesn’t know.

The price→ What it actually sold for not the listed “floor” — listings are easy to fake, sales aren’t

The loan size→ Set by how fast it can sell rarely-traded art means smaller loans, automatically · 35¢ per $1 max (demo)

The exit→ The sale is planned before the loan pre-committed buyers and an auction with a public countdown — no fire-sale improvisation

The honest number→ Bad debt, on the front page a lender that hides its bad-debt rate is marketing · 1.1% (demo)

North star: LTV accuracy vs realised sale price. ★

Loan volume is actively perverse here — it grows fastest when underwriting is worst, and adverse selection guarantees the growth arrives. Every other number inherits its honesty from the oracle’s marks being tested against what liquidated collateral actually fetched.

time-to-liquidate (calm vs stress)bad-debt ratevolume — watched, not targeted