CryptoPunks research / PunkPredictor
Who owns CryptoPunks?
See the largest current holders by address and by curated entity, then follow the evidence into a complete wallet or entity analysis.
Explore current ownership ↓An address is not always a person.
A collector can use several wallets, while a wrapper, vault or escrow address can hold Punks for many people. The address and entity rankings below answer different questions and keep that distinction visible.
Published 2026-09-24. Floor observations through 2026-09-24; native-market events through 2026-09-24 03:59 UTC. Snapshot and methodology.
The public event ledger assigns 10,000 CryptoPunks to 3,977 addresses. These are wallets and custody contracts, not 3,977 distinct people. Canonical public entity records connect 1,936 current holdings across 287 curated entities.
Largest publicly attributed entities
Grouped across current holding addresses using PunkPredictor's canonical, versioned entity records. A label identifies a research grouping; it does not prove legal or beneficial ownership. X profiles appear only when the stored link is verified.
| Entity | Holding addresses | Punks | Collection share | Public profile |
|---|---|---|---|---|
| unrealSov | 2 | 145 | 1.45% | @unrealSov ↗ |
| smithdavid888.eth | 1 | 141 | 1.41% | No verified X profile |
| PunksOTC | 2 | 137 | 1.37% | @punksOTC ↗ |
| Wilcox | 1 | 116 | 1.16% | No verified X profile |
| Mr 703 | 1 | 98 | 0.98% | No verified X profile |
| cool-punks.eth | 1 | 66 | 0.66% | No verified X profile |
| Gary Vee | 1 | 62 | 0.62% | No verified X profile |
| Snowfro | 3 | 53 | 0.53% | No verified X profile |
| DanPolko | 1 | 51 | 0.51% | @DanPolko ↗ |
| Hivemind Digital Culture Fund | 6 | 44 | 0.44% | @HivemindCap ↗ |
| mabu | 1 | 40 | 0.40% | No verified X profile |
| pixls_dot_eth | 2 | 36 | 0.36% | No verified X profile |
| niftynaut | 2 | 33 | 0.33% | @niftynaut ↗ |
| cx000.eth | 1 | 28 | 0.28% | @0x66c7a7 ↗ |
| Kaprekar | 1 | 27 | 0.27% | @Kaprekar ↗ |
| beautyandpunk | 2 | 24 | 0.24% | @beautyandpunk ↗ |
| cyrusthegreat.eth | 2 | 24 | 0.24% | No verified X profile |
| Kanbas | 1 | 24 | 0.24% | @Kanbas_ ↗ |
| shilpixels | 1 | 24 | 0.24% | @shilpixels ↗ |
| 0x650d | 1 | 23 | 0.23% | @0x650d ↗ |
Largest recorded holding addresses
The ten largest custody addresses hold 1,592 Punks (15.92% of the collection). Wrappers, vaults and escrow can represent many beneficial owners, so this is not a richest-people ranking.
| Rank | Entity or public identity | Holding address | Punks | Collection share |
|---|---|---|---|---|
| 1 | No public entity label | 414 | 4.14% | |
| 2 | No public entity label | 222 | 2.22% | |
| 3 | 144 | 1.44% | ||
| 4 | No public entity label | 144 | 1.44% | |
| 5 | smithdavid888.ethNo verified X profile | 141 | 1.41% | |
| 6 | 126 | 1.26% | ||
| 7 | WilcoxNo verified X profile | 116 | 1.16% | |
| 8 | No public entity label | 98 | 0.98% | |
| 9 | Mr 703No verified X profile | 98 | 0.98% | |
| 10 | No public entity label | 89 | 0.89% | |
| 11 | No public entity label | 87 | 0.87% | |
| 12 | No public entity label | 86 | 0.86% | |
| 13 | No public entity label | 72 | 0.72% | |
| 14 | cool-punks.ethNo verified X profile | 66 | 0.66% | |
| 15 | No public entity label | 66 | 0.66% | |
| 16 | Gary VeeNo verified X profile | 62 | 0.62% | |
| 17 | No public entity label | 54 | 0.54% | |
| 18 | 51 | 0.51% | ||
| 19 | No public entity label | 50 | 0.50% | |
| 20 | No public entity label | 50 | 0.50% |
Download ranked addresses with public attribution. Open any address or entity above for its complete wallet analysis.
What entity behavior can explain
Accumulation
When major holders add Punks, supply can tighten in specific cohorts or trait groups.
Distribution
Listings from concentrated holders can change floor pressure and near-floor depth.
Lending activity
Borrowing and refinancing behavior can signal liquidity needs or collateral stress.
Wallet clustering
Grouping related wallets avoids double-counting the same participant's activity.
What PunkPredictor entity intelligence tracks
The app's entity layer joins wallet membership, market-event history, portfolio valuation, loans, and scorecard percentiles. It is meant to answer "who is acting?" without pretending every address is an independent participant.
Wallet membership
Manual entities, algorithmic clusters, direct wallets, proxy wallets, ENS names, social handles, and versioned entity records are resolved into a stable analysis target.
Holdings and value
Current Punk count, featured or most valuable Punk, estimated portfolio value, realized PnL, unrealized PnL, total profit, and claim history.
Market activity
Buys, sales, transfers, internal transfers, bids, cancelled bids, asks created, asks removed, wraps, unwraps, event counts, 30-day activity, and volume.
Lending and risk
Active loans, repaid loans, defaults, current loan exposure, lender and borrower behavior, and collateral context are connected back to entity behavior.
Trader and collector scorecards
Wallet and entity scorecards split market behavior into trader and collector identities. That separation matters because a high-frequency flipper, a long-hold collector, a fund, and a protocol vault can all look active while meaning very different things for the market.
Trader score
Trading identity uses realized profit, win rate, flip ROI, flip time, trade volume, turnover, active trade days, and activity density.
Collector score
Collector identity uses Punk count, hold time, grail count, longest hold, minted Punks, and low lifetime turnover.
Percentile context
Scores become ranks such as top-percentile trader or collector only after comparing against the latest eligible entity population.
Dominant identity
An entity can be more trader-like, more collector-like, or balanced. The public interpretation should preserve that distinction.
How entity grouping changes market interpretation
- Resolve the actor Search can match entity names, tags, handles, ENS names, profiles, or raw wallet addresses before routing to a wallet or entity view.
- Group related wallets carefully A participant using hot wallets, vaults, custody addresses, or proxy wallets should not be counted as multiple unrelated buyers or sellers.
- Separate infrastructure from participants Protocol vaults, marketplace helpers, wrappers, and lending contracts are labeled or excluded where they would pollute trader and collector rankings.
- Read activity through context Listings, bids, loans, transfers, and sales mean different things depending on whether the actor is accumulating, distributing, refinancing, or moving inventory internally.
Leaderboard and actor context
The leaderboard makes aggregate entity behavior scannable: wallets, profit, current holdings, claimed Punks, buys, sales, spent, received, event counts, recent activity, volume, transfers, bids, wraps, asks, and latest event timing. Those fields help explain whether a major holder is accumulating, distributing, financing or moving inventory.
Privacy and publication policy
This public page uses aggregate holder behavior and reviewed entity records. Sensitive, uncertain and algorithmic labels stay out of the crawlable ranking.
How public attribution is handled
Start with aggregates
Wallet clustering, scorecard dimensions and market-event categories explain behavior without inventing an owner for every address.
Require useful context
Address and entity links lead to complete holdings and activity analysis rather than a name attached to a balance.
Mark uncertainty
Manual, algorithmic, protocol, proxy and versioned entities are treated differently throughout the analysis.
Show freshness
Every public ranking carries an observation time and is withheld when the ownership snapshot fails completeness or freshness checks.
Related market intelligence
Lending
Connect holder behavior to collateral, defaults, refinancing, and risk.
Trait prices
See how entities can affect supply in scarce cohorts.
Underpriced listings
Track model-vs-ask gaps that may reflect seller urgency.
Wallet analysis
Open the interactive wallet product for address-level and grouped entity inspection.
CryptoPunks entity FAQ
Why group CryptoPunks wallets into entities?
A collector, fund, trader, or protocol can use multiple wallets. Entity grouping helps interpret accumulation, distribution, liquidity, and lending behavior without treating every address as a separate market participant.
What does PunkPredictor measure for CryptoPunks entities?
PunkPredictor measures wallet membership, Punk holdings, claimed Punks, buys, sales, transfers, bids, asks, wraps, active loans, portfolio value, realized and unrealized PnL, trader score, collector score, and score percentiles where data quality allows.
Why separate protocol entities from collectors and traders?
Protocol vaults, wrappers, lending contracts, and marketplace helpers are infrastructure. PunkPredictor labels or excludes those entities in ranking contexts so they do not compete with human collectors, funds, or traders.
Does PunkPredictor publish private wallet labels?
No. The crawlable ranking uses canonical curated entity records already intended for public analysis. Unreviewed labels and algorithmic clusters are excluded, and X handles appear only when their public link is verified.
Wallets, entities and public owners
A wallet is an address. An entity is a grouping used to interpret related wallets. A publicly documented owner is a person or organization whose ownership has been disclosed in a reliable public source. These are three different kinds of evidence.
A documented example: Visa
Visa documented purchasing a CryptoPunk in August 2021 in its metaverse research paper. That establishes a historical purchase; it is not, by itself, evidence of a current holding. We do not label the current address ranking with celebrity names inferred from wallet movements.
Why the largest address may not be the largest collector
A wrapper or vault can hold Punks for many owners. A collector can also use several wallets. The two rankings above therefore answer different questions: one groups current custody addresses through canonical public entity records, while the other preserves the reproducible address-level ledger. Follow either kind of row into PunkPredictor for deeper holdings and activity analysis.