Prestige
The $N 000 -$100 k drop
Small edition (5–50), high price (1–10 ETH per piece). You are selling scarcity of selection. The audience is established collectors. Marketing is one-to-one.
Sell → Pricing
Price is the most consequential decision in a generative drop — more important than platform choice, more important than marketing window. Get the price right and the algorithm does the selling. Get it wrong and you burn the audience's confidence in your future work.
Four knobs to set, three of which are decisions you can change per drop. The fourth — your reputation — accumulates across them.
The combination of those four numbers is your drop's identity. A 100-edition at 2 ETH with 100% curation says "every output is worthy." A 1 000-edition at 0.05 ETH with 40% curation says "the algorithm is the work; I curate the highlight reel." Neither is more correct than the other; they address different audiences.
Prestige
Small edition (5–50), high price (1–10 ETH per piece). You are selling scarcity of selection. The audience is established collectors. Marketing is one-to-one.
Volume
Large edition (500–10 000), low price (0.01–0.1 ETH). You are selling breadth of reach. The audience is everyone. Marketing is one-to-many.
Most artists hit one of these on their first drop and never leave. The mistake is to assume the model you start with is the one you keep. A practice that starts with $100-floor editions can graduate to $5k 1/1 auctions once the audience is established — and vice versa, an auction-led practice can use $100-floor editions to onboard new collectors.
At 1 ETH ≈ $3 500 (mid-2026):
| Model | Edition | Price | Gross primary | After 10% fee | Per-collector economics |
|---|---|---|---|---|---|
| Auction 1/1 | 1 | 10 ETH | 10 ETH | 9 ETH | One collector, high prestige |
| Curated 100 | 100 | 2 ETH | 200 ETH | 180 ETH | Each collector owns a curated piece |
| Volume 1 000 | 1 000 | 0.05 ETH | 50 ETH | 45 ETH | Low entry; broad base |
| Open mint | 10 000 | 0.01 ETH | 100 ETH | 90 ETH | Liquidity play; secondary is the story |
The 100-edition at 2 ETH generates more revenue than the 10 000 at 0.01 ETH — but the open mint builds a collector base 100× larger, which compounds on secondary royalties. Neither is strictly better; they compound differently.
Your algorithm produces 4 096 distinguishable outputs (a 12-bit seed space). Three pricing strategies, with realistic numbers.
Pick the seed that produced the most striking output (your "hero"), render it at 4K, and auction it as a 1/1. List at 10 ETH. The 4 095 other outputs you keep on the shelf — available for a future drop, or never released.
Run the algorithm, curate 100 of the 4 096 outputs (the top 2.5% by your taste), render each, mint as an edition of 100. Burn the other 3 996 seeds (or hold them as reserve).
Mint all 4 096 outputs, with the algorithm on chain and each seed generating the artwork at mint time. Price at 0.05 ETH per piece.
| Strategy | Revenue at 100% sell-through | Revenue at 50% sell-through | Audience impact | Reputation impact |
|---|---|---|---|---|
| A — 1/1 auction | 10 ETH | 10 ETH (only 1 to sell) | Negligible | High prestige, no compounding |
| B — curated 100 | 200 ETH | 100 ETH | Medium (100 new holders) | Strong — collectors see curation as the value |
| C — algorithmic 4 096 | 205 ETH | 103 ETH | Large (up to 4 096 holders) | Depends on curation rate and tail quality |
Options B and C have similar gross revenue at full sell-through, but the algorithmic drop (C) compounds harder on secondary — because there are 40× more tokens in the world, each paying you a royalty. See the royalty section below.
Drops rarely sell out at the price you wanted. The question isn't whether there will be unsold inventory — it's what you do with it.
0x000...dead. Floor is established by the
sold quantity alone. This is what Tyler Hobbs did with the
unsold end of the Fidenza supply.
The math on each:
| Strategy | Outcome at 60% sell-through | Best when | Risk |
|---|---|---|---|
| Burn | Edition reduces to 60% of planned size; floor anchored at primary price. | The piece is strong and the audience values scarcity. | None — irreversible and clean. |
| Release at floor | Floor slowly drifts toward primary as remaining inventory clears. | You can wait months for the right buyers. | Floor drift; secondary liquidates at lower price. |
| Roll forward | Inventory held in artist's wallet; re-mint window set. | You have a planned follow-on drop and audience growth. | Holding cost; perception of "unsold." |
Don't burn quietly
Whatever you do with unsold inventory, announce it on-chain and on social channels the day the mint closes. "We burnt 1 234 unsold tokens" is good news. "We silently released them at floor six months later" is the kind of disclosure that loses collector trust.
Secondary sales are where generative-art revenue compounds. ERC-2981 is the on-chain standard for declaring royalties.
The contract exposes a royaltyInfo(tokenId, salePrice)
function that returns a recipient address and a basis-points cut
(e.g. 500 = 5%). Marketplaces (OpenSea, Blur, Magic Eden,
LooksRare, etc.) honour it automatically.
The artist receives the cut on every secondary sale, in the currency of that sale, with no manual claim required. Some marketplaces (especially on Solana and Tezos) use their own royalty mechanism; on Ethereum mainnet, ERC-2981 is dominant.
| Rate | Pros | Cons | Audience signal |
|---|---|---|---|
| 0% | Maximises secondary liquidity (no friction for traders). | No passive income from the drop. | "Trader-friendly." |
| 2.5% | Standard for many 2024–26 drops; balances income with liquidity. | Below industry norm for generative. | "Fair." |
| 5% | Industry standard for generative art (Art Blocks default). | Some marketplaces apply lower; a few traders avoid. | "Aligned with artists." |
| 7.5–10% | Strong passive revenue; signals commitment to the practice. | Higher friction; can reduce volume. | "Long-term artist." |
| > 10% | Defensive against wash trading for inflated volume. | Most marketplaces cap effective enforcement; rare. | "Reserved for blue-chip 1/1s." |
For a first drop, 5% is the safe default. It matches what Art Blocks charges, what Manifold defaults to, and what most collectors expect from generative work. If your collector base is purely trader-driven (you can tell by the wallet distribution), 2.5% may move secondary volume meaningfully. If your collectors are long-term aligned (holders from previous drops), 7.5% compounds cleanly.
A 5% royalty on a 100-ETH secondary sale is 5 ETH. A 5% royalty on 50 small resales of 0.5 ETH each is 1.25 ETH total. Volume compounds; price per sale compounds. The combination — how many sales, at what price — is what determines the secondary income over the life of the collection.
Practical tip
Set the royalty at deploy time. Changing it later requires a contract migration or a Manifold extension update, both of which are noticeable on-chain and reduce collector confidence.
Most generative artists never sell the script. A few do.
Some practices license the algorithm to other artists for a flat fee + per-mint royalty. This works when:
Realistic licensing numbers for a proven generative system (mid-2026):
| License type | Typical fee | Royalty | Exclusivity |
|---|---|---|---|
| Non-exclusive | 5–25 ETH upfront | 2.5% of licensee's primary sales | None |
| Time-limited exclusive | 25–100 ETH upfront | 5% of primary sales | 12–24 months |
| Full buyout | 100–500+ ETH | None | Permanent |
Full buyouts are rare in generative — the algorithm is the artist's practice, not a product. They happen when an artist exits the space or when a brand wants to anchor a campaign. Non-exclusive licenses are more common and tend to align both sides' interests.
This is not legal advice
License terms are a contract, and a poorly drafted one can outlive the algorithm's relevance. Use a lawyer. The numbers above are market references, not templates.
Next: the legal and tax basics at /sell/legal/, or pick a platform at /sell/platforms/.