Stablecoin Map Methodology: Data Sources, Supply and Peg Definitions

Scope and coverage

Objective

The Global Stablecoin Map measures the supply, the peg and the distribution of active stablecoins across the chains DIA reads. It is a data reference. It publishes measurements and assigns no ratings. This document states how each published figure is produced, when it is withheld and how it is revised, following the contents a benchmark methodology is expected to carry^[1].

Universe and inclusion criteria

The map spans the dollar, the euro, sterling and 22 other currencies. The counts in this document come from the live registry, the same one behind the homepage counts. The rules below decide what is listed and what is counted.

A stablecoin is listed when its status is active.
Supply is only ever measured from contract deployments DIA has verified onchain.
3 wrapped stablecoins, such as sDAI, are listed but not counted in any total or count on the map.
A listed stablecoin is counted when it has measured supply: a dollar value above zero, or a first-party total withheld by the coverage rule in section 4.1.2, whose supply exists but is not shown today. Every count on the map uses this population.
1 listed stablecoin has no measured supply yet: it stays listed, marked “Supply not read” when no contract was read or “Supply not priced” when tokens were read but no price exists, and is left out of every count.
An issuer or a peg currency is counted when at least one of its stablecoins is counted.
A chain is counted when listed stablecoins hold $100K+ on it between them, valued as in section 4.1.6. Smaller chains stay listed on the chain page.
A peer comparison on a coin page draws from counted stablecoins only, and each card names the population it ranks against.
Only active stablecoins appear in the ranked tables and in the currency, region and issuer pages.
A stablecoin that has wound down keeps its own page, marked historical, and is excluded from those tables.

Metadata at a glance

Statistical unit: One stablecoin: one token, counted once across all of its chains.
Population: 143 stablecoins with measured supply, from 99 issuers.
Reference area: 72 chains with $100K+ of supply. 25 peg currencies.
Unit of measure: US dollars for value, native token units for supply, basis points for peg deviation, percent for shares.
Reference period: The snapshot taken at each update. Charts keep one point per UTC day: the last update of that day.
Frequency: 4 a day, in runs that start at 03:00, 09:00, 15:00 and 21:00 UTC (section 7.5).
Time coverage: From each stablecoin's first DIA supply reading. Supply charts show DeFiLlama history before that day (section 4.1.5).

Coverage at a glance

143Stablecoins with supply

99Issuers

25Peg currencies

72Chains with $100K+

4 a dayUpdates

48 hoursOldest price used

Definitions and classification

Glossary

Active stablecoin: A stablecoin whose status is active. Only these enter tables and totals; counts take the counted ones. Section 1.2.
Basis point (bp): One hundredth of a percent, 0.01%.
Bridged leg: A leg on one chain that is a claim on tokens already counted on another chain. Section 4.1.1.
Circulating supply: Contract supply less the treasury, premint and burn balances, published only when that list is complete and every balance read. Formula (4.2).
Contract supply: The token total read from the verified contracts that read in a run. Formula (4.1).
Counted stablecoin: A listed stablecoin with measured supply: a dollar value above zero, or a withheld first-party total. Every count on the map uses these. Section 1.2.
Excluded role: A treasury, premint or burn address, whose balance is taken out of circulating supply. Section 4.1.
First-party figure: A figure computed from DIA's own contract reads and DIA's own price. A DeFiLlama figure is labelled as not measured by DIA. Section 3.3.
Headline holder chain: The one chain on which a stablecoin's holders are counted. Section 4.5.
Herfindahl-Hirschman index (HHI): The sum of the squared holder shares in percent, from 0 to 10,000. Formula (4.12).
Holder: An address with a balance above zero on the headline holder chain, frozen addresses included. Section 4.5.
Leg: One stablecoin's supply on one chain.
Lending share: The share of a stablecoin's value supplied to the lending markets DIA reads. Formula (4.9).
Peg currency: The currency a stablecoin is designed to be worth one unit of.
Peg deviation: The distance between a stablecoin's price and one unit of its peg currency, in basis points. Formula (4.6).
Supply not read, Supply not priced: The marks on a listed stablecoin with no measured supply: no contract was read, or tokens were read but no price exists. Such a coin is left out of every count. Section 1.2.
Wrapper: A token that holds another listed stablecoin, such as sDAI. It is listed and left out of every total and count. Section 1.2.

Issuer types and licences

Each stablecoin is listed under the company or protocol that issues it. Each issuer has a form and a licence tier. The rules below run top to bottom, and the first match decides.

When a stablecoin is sold under another company's brand, that company is recorded as its brand owner. The brand does not change the issuer's form. The stablecoin page names both.

Issuer form

What kind of legal thing issues it?

1. Bank
The issuer, or a parent that controls it, holds a deposit-taking banking licence on a regulator's register.
2. Trust company
The issuer holds a trust charter with no deposit-taking: NYDFS limited purpose trust, OCC national trust bank, US state trust company or Canadian provincial trust corporation.
3. Non-bank company
A named legal entity mints and redeems (including through a KYC allowlist), or holds or custodies the reserve off-chain.
4. Protocol
Otherwise: minting and redemption are open to any address meeting onchain conditions, and the reserve sits in smart contracts.

Licence tier

Is it authorised to issue, and where? An issuer takes the highest tier that any of its licences or registrations reaches.

1. Recognised regime
Holds an authorisation covering issuance in a recognised regime (listed below).
2. Licensed elsewhere
Holds an issuance licence from a regulator outside the recognised regimes.
3. Registered elsewhere
Registered as an issuer, or its token or fund registered, with a regulator outside the recognised regimes, and holds no issuance licence.
4. Registration only
Holds only an AML or VASP registration, SRO membership, sandbox or incubation status, a pending application, a securities exemption or a MiCA white paper.
5. None found
We checked the registers and the issuer's own pages and found no licence or registration. Includes protocols with no issuing entity.
6. Not checked
No regime on record: we have not looked for a licence for this issuer yet.

Recognised regimes

A licence puts an issuer in the Recognised regime tier only when it covers issuance and was granted in one of these:

EU and EEA under MiCA (e-money institution or credit institution)
United States (state money transmitter, NYDFS, OCC, GENIUS Act permitted issuer)
United Kingdom (FCA e-money)
Switzerland (FINMA licence, not self-regulatory organisation membership)
Singapore (MAS)
Hong Kong (HKMA)
Japan (JFSA)
UAE (CBUAE, VARA, ADGM, DIFC)
Australia (ASIC AFSL)
Canada (CSA value-referenced crypto asset framework)
The regulator of any jurisdiction whose central bank issues a currency the issuer's coins track

An issuance licence from any other regulator, including El Salvador, Bermuda and the Isle of Man, puts the issuer in Licensed elsewhere.

Stablecoin layers

Each stablecoin gets one value in each of three layers. Each layer answers one question. The rules run top to bottom, and the first match decides. For a wider taxonomy of stablecoin designs, see ^[2].

Stablecoin layers and their values, in evaluation order
Layer	Question	Values, in evaluation order
Backing	What is the coin designed to be backed by?	Own-token / algorithmic. There is no collateral, or the collateral is mainly the issuing protocol's own token. Cash & money-market. Designed to be backed by bank deposits, cash, government bills of a year or less, reverse repo or money-market fund shares (looked through for funds only). Other stablecoins. Designed to be backed by other stablecoins or lending receipts on them, with no look-through. Crypto collateral. Designed to be backed by volatile crypto or staking tokens held as overcollateral, unhedged. Hedged crypto. Designed to be backed by crypto whose price exposure is offset by derivative shorts, spot and hedge counted as one position. Credit & other RWA. Designed to be backed by loans, private credit, corporate paper, government debt over a year, or other real-world assets.
Holder return	Does simply holding it pay you?	Price accrues. An ordinary holder sees a unit's redemption value rise by design (share price, NAV accrual, CPI index). Balance rebases. An ordinary holder's balance grows without any action (rebase, or dividends paid in new tokens). None. Holding pays nothing, including coins whose yield goes only to allowlisted earners or to a separate staked token.
Holder recourse	Who, if anyone, owes you a unit of the peg?	Issuer redemption. A named entity's terms give holders a right to redeem 1:1 at that entity, possibly after onboarding or KYC. Contract only. Redemption runs only through contracts, pools or liquidations; no entity owes the holder a unit of the peg.

Data sources

Source inventory

Primary sources are read by DIA. Reference sources check or complete them. The aggregator, DeFiLlama^[3], stands in only where DIA has no figure. Run times are UTC.

Sources, the metrics they feed, run times and freshness rules
Source	Role	Metrics fed	Runs (UTC)	Freshness rule
On-Chain Contract Reads	Primary	Supply per chain; treasury, premint and burn balances	03:00, 09:00, 15:00, 21:00	A failed chain is a failure, never zero
DIA Price Aggregation	Primary	Price, peg deviation	With each supply update	48 hours; symbol-matched quotes rejected
Fiat Reference Rates	Reference	Dollar rate of non-dollar peg currencies	With each supply update; ECB on business days	No quote, no peg deviation
Peg Cross-Check	Reference	DeFiLlama and CoinGecko current prices	With each supply update	Section 4.3.3
Block Explorers	Reference	Holder counts, top holders	With each supply update	Section 4.5
Lending Markets	Reference	Supplied balances, utilisation, supply rates	00:00, 06:00, 12:00, 18:00	Markets not updated for 3 days left out
DeFiLlama Fallback	Aggregator	Supply with no DIA figure, chain split, history before DIA's first read	00:30, 06:30, 12:30, 18:30	Labelled as not measured by DIA
Price history	Reference	Peg history charts	One close per day	DeFiLlama, or CoinGecko where DeFiLlama has none
Contract controls	Primary	Pause, freeze and upgrade functions	Sundays, 05:00	Section 4.6
Issuer Research	Manual	Identity, jurisdiction, licence, reserves, transparency links	On review of a filing, licence or reserve report	Each row carries its documents

Source descriptions

On-Chain Contract Reads

Primary

Supply is read from the token contract on each chain where DIA has verified a deployment, and summed across chains. A chain that fails to read is recorded as a failure and never counted as zero. A coin with no successful read reports no supply.

Refresh4 times a day

DIA Price Aggregation

Primary

Prices come from DIA's own aggregation across trading venues. A quotation matched by symbol instead of by the specific contract is rejected. A price older than 48 hours is never used.

Refresh4 times a day, prices older than 48 hours not used

Peg Cross-Check

Reference

Before a peg deviation is shown, DIA's price is compared with the current price from DeFiLlama and from CoinGecko. The rule is in section 4.3.3.

Refresh4 times a day

Fiat Reference Rates

Reference

A non-dollar stablecoin needs its currency's dollar rate before its peg deviation can be measured. The euro and sterling come from DIA's own fiat feed. Other currencies use the ECB reference rate, or open.er-api.com where the ECB has none. A currency that no source quotes shows no peg deviation.

Refresh4 times a day, ECB rates change on business days

Block Explorers

Reference

Holder counts and top holders come from block explorers such as Blockscout and Etherscan. The headline holder count is taken on one chain per stablecoin.

Refresh4 times a day

Lending Markets

Reference

Balances deposited in the lending markets DIA reads for its vault map. The stablecoin pages show how much of each stablecoin sits in them. A market not updated for three days is left out.

Refresh4 times a day

Issuer Research

Manual

Issuer identity, jurisdiction, licensing, reserve posture and transparency links are researched by hand against primary sources: issuer filings, regulator registers and published attestations. Every researched row carries the documents it was built from. Where a coin is brand-owned but issued by a licensed platform, both entities are recorded. The issuer owes the holder, not the brand.

RefreshManual review

DeFiLlama Fallback

Aggregator

Coins with no DIA figure yet fall back to DeFiLlama for the supply figure and are not labelled as measured by DIA.

Refresh4 times a day

Source hierarchy

DIA's own contract reads and DIA's own price come first. A figure built from them is a first-party figure.
Where DIA has no supply figure or no price for a stablecoin, DeFiLlama's circulating figure stands in, and the figure is labelled as not measured by DIA.
For a stablecoin no market source quotes, a curated exchange rate, or for a staked version the vault's share price times the price of the coin it holds, can value the supply. Such a price carries no peg reading.
Sources are never mixed inside one figure without a label. A total DIA withholds is not replaced with DeFiLlama's figure.

Calculation methods

Each subsection gives the definition, the formula and the conditions under which the figure is withheld. Formulas are numbered (4.1) to (4.12) for citation. A withheld figure is shown as withheld. It is never replaced by an estimate.

Circulating supply

Contract supply is the token total read on chain. Circulating supply is contract supply less the balances nobody holds: the issuer's own treasury wallets, preminted supply the issuer has not released, and burn addresses. Tokens a custodian or a bridge holds for users stay in. Free-float supply measures take the same approach of subtracting balances unlikely to trade^[4].

S_{contract} = \sum_{c \in C} (T_{c} - K_{c})

(4.1)

where T_c is the token's totalSupply() on chain c, C is the set of verified chains that read in the run, less bridged legs (4.1.1), and K_c is the balance of a mapped bridge escrow on chain c, such as the Fraxtal bridge escrow on Ethereum for FRAX and frxUSD, and 0 on a chain with none. A chain whose escrow balance fails to read, or exceeds the chain's own supply, is treated as a failed read and left out of C.

S_{circ} = S_{contract} - \sum_{a \in E} b_{a}

(4.2)

where E is the curated list of treasury, premint and burn addresses on the chains in C, and b_a is the balance of address a. An address on a chain outside C is not subtracted.

Withheld when

the coin's exclusion list is not curated, cited and marked complete;
any balance in E fails to read;
the result is below zero or above S_contract.

In each case the headline is contract supply. Where circulating supply publishes, issuer-held supply is S_contract minus S_circ.

One exception: crvUSD is issued as debt, so its circulating supply follows Curve's own definition, the crvUSD borrowed and not repaid across its mint markets, peg keepers and Yield Basis markets, read from the contracts in place of (4.2). One unreadable part withholds the figure. Stored circulating supply for crvUSD counts from 2026-10-05, the first day on this definition.

Bridged legs

A bridged leg is a claim on tokens already counted on another chain, so adding it to the source chain would count the same dollars twice. Where the bridge is mapped, DIA counts the bridged amount on the chain that holds it and takes it off the chain that backs it, so the total counts each dollar once. Only a coin with bridged legs nobody has mapped has its first-party total withheld. In both cases the per-chain legs stay correct and are still rendered separately. Where one escrow on the source chain backs only chains DIA reads, its balance is subtracted from the source chain instead, the K_c term in (4.1).

Withholding rules

Some totals are withheld even when every chain read is correct, because the sum would misstate the coin. A smaller unissued overhang still publishes the figure, labelled with the reason it overstates circulation.

Rules that withhold or label a supply total
Rule	Applies when	Effect
Unmapped bridged legs	Chains hold bridged claims on tokens counted on another chain, and the bridge is not mapped	Total and DIA price withheld; legs still shown
Mint and burn cycling	The contract mints and burns on a cycle, so one read is a point-in-time balance	Total and DIA price withheld
Dominant unissued supply	Almost all of the onchain supply is unissued	Total and DIA price withheld
Concentrated premint	The holder scan finds 10 or fewer holders and 95% or more in one address	Total and DIA price withheld
Coverage regression	A chain read in at least 3 of the last 5 published snapshots is missing from this run	Total withheld until the chain reads again or leaves the window
Outlier leg	Section 4.1.3	Total withheld; the outlier leg is dropped
Unissued overhang	The issuer states that some onchain supply is not in circulation	Published, labelled unissued
Premint concentration	Curated: almost the whole supply sits in a handful of wallets and has barely moved	Published, labelled concentrated

Outlier-leg guard

With three or more chains read, the total is withheld when the largest leg holds more than 1,000,000 tokens and more than 1,000 times the second-largest leg, and the second-largest leg is above zero. Such a leg is a decimals or minting error, not a concentration of the same token. A coin labelled with an unissued overhang or a premint concentration publishes its total. A coin whose record names its home chain accepts a home-chain leg up to 100,000 times the second-largest.

Frozen and destroyed funds

Some issuers can freeze an address, and a few can also destroy the balance of a frozen address. We count them the same way for every coin. Supply is the onchain total supply, so frozen tokens stay in supply and in circulating supply: the issuer still owes them and can release them. When an issuer destroys a frozen balance, the tokens leave supply at the block where it happens. Holders are the addresses with a balance above zero, frozen addresses included. Holder history is rebuilt from every event that changes a balance, including blacklist wipes that emit no transfer. When we cannot verify that a coin's history reproduces its supply exactly, we do not publish that history. Frozen stablecoin funds lists the current balances and which contracts can freeze and destroy.

Value, fallback and history

V = S \times P

(4.3)

where S is the published supply (circulating where it publishes, contract supply otherwise) and P is the published price in US dollars.

Where no first-party supply or no price exists, V is DeFiLlama's circulating dollar figure, labelled as not measured by DIA. A total withheld under section 4.1.2 has no value.
The supply chart uses DeFiLlama history only for days before DIA's first supply reading. A day is never drawn from both sources. A step of more than 3% at the junction is marked on the chart. When the two series differ by more than 5 times at the junction, the DeFiLlama history is dropped.
The supply chart draws DIA’s contract supply, with DeFiLlama’s figure before DIA’s first day. Circulating supply is stored for only a short part of the window, so it is the headline figure and is not charted.

Per-chain split

With a DIA price, each chain's value is its token count times the price, and the chains are scaled down together so they never add up to more than the coin's value. Without a DIA price, DeFiLlama's total is split in the proportions of DIA's token counts. Where a chain holding at least 1% of the coin differs from DeFiLlama's share for that chain by more than 25%, DeFiLlama's own split is used.

The Supply by chain rows on a coin page add up to the headline.

Each issuer treasury comes off the chain that holds it.
Bridged supply shows on the chain that holds it and comes off its source chain. A bridged leg that cannot be placed, because its source chain is not counted or has nothing left, lists its read and stays out of the sum.
Where the rows carry more than the headline, as with a bridge escrow or a treasury with no per-chain split, the difference is one Not circulating row, and shares are of the chain rows. Where the headline is above the rows, the difference is one Not split by chain row.
No shares are shown when the total is withheld or is not a first-party figure.

Market aggregates

V_{total} = \sum_{i \in A} V_{i}

(4.4)

where A is the set of active stablecoins, wrappers out, that have a value V_i from (4.3).

w_{k} = \frac{\sum_{i \in A_{k}} V_{i}}{V_{total}}

(4.5)

where A_k is the subset of A pegged to currency k.

Supply by chain is the sum of each coin's chain values from section 4.1.6. The home page shows the 10 largest chains.
crvUSD enters every total at its debt-based circulating supply (section 4.1).
A stablecoin counts once in each region it is tagged with, so region figures overlap and do not add up to the total.

Peg deviation

Peg deviation measures a price against par, one unit of the peg currency^[5].

d = round (10^{4} \times (\frac{P_{USD}}{R_{peg}} - 1))

(4.6)

where d is in basis points, negative below the peg, P_USD is DIA's price in US dollars, and R_peg is the dollar value of one unit of the peg currency: 1 for the dollar, DIA's fiat feed for the euro and sterling, and the ECB reference rate^[6] or open.er-api.com for other currencies.

Eligibility

Peg deviation is computed only for stablecoins designed to hold exactly one unit of their peg currency, so yield-bearing coins that appreciate by design are not reported as depegged. A coin whose holder return accrues into its price gets no reading. A rebasing coin holds one unit while its balance grows, so it is measured like any other. A price set from an exchange rate or a vault share price carries no reading, and a currency that no source quotes shows no peg deviation.

Freshness

A price older than 48 hours is never used. The peg figures on the home page use only readings from the latest update day or the day before.

Cross-check

Before a peg deviation is shown, DIA's reading is compared with the readings from the current DeFiLlama and CoinGecko^[7] prices.

| d_{DIA} - d_{r} | \leq \max (τ, 0.15 \times \max (| d_{DIA} |, | d_{r} |))

(4.7)

where d_r is the deviation from (4.6) at reference r, DeFiLlama or CoinGecko, and τ is 50 bp, or 150 bp for yen coins.

The figure shows when (4.7) holds for at least one reference that has a price. It is withheld when DIA disagrees with every reference that has a price.
A DeFiLlama reading more than 1,000 bp from both DIA and CoinGecko is treated as a failed source and ignored.
With no reference price, DIA's reading stands as unchecked. A DIA price within 1 bp of a daily fallback exchange rate is then treated as that rate, not a market price, and withheld.
A disputed DIA price is hidden as well when the previous run published no peg reading for the coin, or when the price is identical to the previous run's price.

Home-page figures

Today's peg figures count stablecoins with a current reading and at least $1M of supply, wrappers out. The headline share of supply within 50 bp of $1 and the distribution in bands at 10, 50 and 100 bp cover dollar coins.
The same readings are grouped by peg currency: dollar, euro, and every other currency together. Each group reports how many coins sit within 10, 50 and 100 bp, and the median deviation. A non-dollar reading is DIA's dollar price divided by the day's exchange rate, so it carries that rate's timing.
A 30-day panel lists, for dollar coins, the closes furthest from $1 over the last 30 days and how many coins closed more than 100 bp away. It reads the stored daily closes from DeFiLlama and CoinGecko, so a coin with no reading today can still appear.
A deviation of more than 500 bp carries one sentence with the date of the reading and what DeFiLlama shows for the coin.

Peg stability over time

Coin pages describe the peg over time from one daily close per coin, taken from DeFiLlama, or CoinGecko where DeFiLlama has none. The daily deviation is d_t = 10⁴ × (p_t − 1), where p_t is the close in the coin's own peg currency. Coins whose holder return accrues into the price are not measured.

σ = \sqrt{\frac{1}{n} \sum_{t \in W} {(d_{t} - \bar{d})}^{2}}

(4.8)

where W is the 90 calendar days ending on the last close (30 days for the short figure), n the number of closes in it and d̄ their mean. With fewer than 2 closes there is no figure.

Largest move: the close furthest from par in the last 12 months, with its date and sign.
Peer ranking: where the coin's 90-day volatility sits among the active, non-accruing coins of its peg currency with a 90-day figure, the coin included. It is shown only when at least 3 coins have one. The peer chart and table cover dollar coins only.
Exchange-rate timing: a non-dollar close is a dollar price divided by an exchange rate taken at a different moment, so a currency move between the two prints the same false move on every coin of that currency. A day is left out of a non-dollar coin's largest move when all of these hold: the coin is at least 50 bp from par; at least 3 other coins of the same peg currency closed that day; a strict majority of them sit on the same side of par within max(25 bp, 35% of the coin's deviation) of it; and their median is no more than 200 bp from par. The page says how many days were left out.
Left out as provider errors: a single close more than 2,000 bp from both neighbouring days when those two are within 500 bp of each other; filled closes of exactly 1.0000 before 2025-10-01; MXNT closes from 2025-04-01 to 2026-01-22, priced from a thin trading pool; and a short list of dated prints. Stored data is not changed, and the chart notes each exclusion.

The share of a coin's onchain supply that sits in lending markets DIA reads.

L_{i} = \min (1, \frac{M_{i}}{S_{i} \times P_{i}})

(4.9)

where M_i is the dollar value supplied to the counted markets, S_i the onchain supply and P_i DIA's price, or a fallback market price between 0.01 and 10 times the peg.

In: markets that report a balance and an update within 3 days.
Out: any market holding less than $1,000 of the coin; abandoned markets whose balance grows only from accrued interest (utilisation of 99.9% or more with a supply rate of 50% or more); DEX liquidity, CDP collateral and isolated collateral.
Rebasing coins show no share. The figure is a floor, since chains and protocols DIA does not read are not counted.

L = \frac{\sum_{i} \min (M_{i} - M_{i}^{self}, V_{i})}{\sum_{i} V_{i}}

(4.10)

the home-page share across stablecoins, where M_i^self is the issuer's own supply of its stablecoin to those markets and V_i is from (4.3). Wrappers and rebasing coins are out of both sums.

The share on loan is borrowed balance over supplied balance, over the markets that report utilisation. It is shown only when those markets cover at least 50% of the lent balance. A coin page's average supply rate and utilisation follow the same 50% rule.

Holders and concentration

A holder is an address with a balance above zero, frozen addresses included.
Holders are counted on one chain per stablecoin: the largest chain an explorer can read, which must exceed the default chain, Ethereum where the coin has it, by more than 5%. The default chain keeps a tie. When that chain fails to read, the next largest chain holding at least 10% of the readable supply is used.

s_{N} = 100 \times \frac{\sum_{j = 1}^{N} b_{(j)}}{S_{h}}

(4.11)

where b_(j) is the j-th largest balance, N is 1, 5, 10, 50 or 100, and S_h is the coin's verified supply on that chain.

HHI = \sum_{j} {(100 \times \frac{b_{j}}{S_{h}})}^{2}

(4.12)

over every holder j, on a scale of 0 to 10,000.

A top-N figure is published only when the holder list read reaches N addresses or covers every holder, and is left blank otherwise. HHI is published only when the read covers every holder.
No share is published when the balances read add up to more than the chain's supply.
What is left out depends on what the source returns, and the coin page names the method under its table:
- Net of contracts and treasuries (Blockscout, and Ethplorer, which keeps Blockscout's shares): the issuer's treasury wallets and burn addresses leave the holder list and the supply; addresses named as protocol or bridge contracts leave the holder list but keep their supply in the denominator.
- Net of treasuries (Tronscan, NodeReal and DIA's own replay of transfers): treasuries and burn addresses are netted the same way. These sources name no contracts, so contracts count as holders.
- Gross (Routescan and the Solana cluster): these sources return no addresses, so nothing is netted and shares are of the whole supply on the chain.
Exchange wallets always stay in.
The map publishes these measures and attaches no labels to them.

Contract controls

Once a week, on Sundays, DIA records the pause, freeze and upgrade functions found in each token contract, on EVM chains, Kava included, and on Tron, Aptos, NEAR and TON. A control is marked absent only when a verified source or the chain state shows it. Frozen stablecoin funds lists the contracts that can freeze and destroy balances.

Data quality and validation

Failure handling

A chain that fails to read is a failure, never a zero. A coin with no successful read reports no supply.
A withheld figure is never replaced by an estimate.
One unreadable treasury, premint or burn balance withholds circulating supply (section 4.1).
An unread holder count is not a zero count. A failed holder scan never withholds supply on its own.
A lending read that returns nothing shows no lending figure, not 0%.

Price cross-check

Every peg reading is checked against two independent prices before it is shown. The rule is in section 4.3.3.

Independent reconciliation

Each day, every published total is compared with an independent directory. For each pair the relative gap is

δ = \frac{| x_{ours} - x_{ref} |}{\max (x_{ours}, x_{ref})}

(5.1)

where x_ours is DIA's figure and x_ref the reference figure, both in US dollars.

Each pair takes the first severity that applies:

missing ours: our figure is below the floor and the reference is at or above the missing threshold;
missing reference: the reference is below the floor and ours is at or above the missing threshold;
ok: the absolute gap is below the floor;
large: δ is at or above the large percentage and the absolute gap at or above the large amount;
minor: δ is at or above the minor percentage;
ok otherwise.

Reconciliation tolerances by lane
Lane	Reference	Minor	Large	Floor	Large amount	Missing threshold
Supply	DeFiLlama stablecoins	5%	20%	$1M	$25M	$10M
Lending	DeFiLlama lending, Morpho	15%	40%	$0.5M	$10M	$5M
Supply per chain	DeFiLlama stablecoins, per chain	5%	10%	$1M	$50M	$50M

Further checks run in their own units:

Reconciliation checks with their own thresholds
Check	Compares	Minor	Large	Scope
Peg today	DIA peg deviation against CoinGecko's	25 bp	75 bp	A missing reading is raised for the 30 largest coins only
Peg volatility	90-day volatility of the daily deviation, recomputed	10 bp	25 bp	The 20 largest coins
Unchecked share	Share of dollar coins with no published peg reading	Not used	Above 20%	Coins with a snapshot in the last 7 days
Overhang	Published supply more than 25% above DeFiLlama's token count, with 70% or more in the top 5 holders	Not used	Both conditions hold	Excess of $1M or more
Backing classes	Cash and money-market total against DeFiLlama's fiat-backed total	A coin sits in only one of the two	More than 2% apart	Coins matched in both
Holder counts	Stored holder count against a second explorer read	5%	20% and 1,000 holders	Every 7 days, the 20 largest coins on EVM chains
Yields	Shown pool rates against DIA's onchain lending rate	1.0 points and 25%	3.0 points and 50%	Pools shown in the last 14 days with $5M or more deposited

A large or missing-ours gap opens a review, raised once per streak. A streak stays open while the subject stays at that severity within 7 days. A gap explained by a known difference in definition, such as a labelled supply caveat, is kept and shown as minor with the reason.

Reader freshness monitoring

Silent: a reader with no successful write for 24 hours.
Dropped: a run that writes less than half the median of the last 7 days, where there are at least 4 earlier runs and the median is 5 rows or more.
Never written: a registered reader with 3 or more attempts and no rows. A lending or supply reader with no rows at all alerts on its first attempt.
Stale markets: a live reader whose markets holding $1M or more have not been written for 3 days. Morpho markets are left out, because the Morpho reader drops abandoned and dust markets on purpose.

Manual research controls

Issuer records use primary sources only: issuer filings, regulator registers and published attestations. Each researched row carries the documents it was built from. The brand owner and the issuer are recorded separately.

Limitations

The figures have known limits^[1]:

Holder counts and concentration cover one chain per stablecoin.
HHI and the top 50 and top 100 shares need a complete holder read, so many coins show none.
Peg history charts come from DeFiLlama and CoinGecko closing prices, not DIA's price feed.
Figures from the DeFiLlama fallback are not first-party.
ECB reference rates move on business days only.
A non-dollar peg reading divides a dollar price by an exchange rate taken at a different moment, so it carries the timing gap between the two (section 4.3.5).
A currency that no source quotes shows no peg figure.
Issuer research is updated by hand and can lag a new filing.
Lending figures cover only the markets DIA reads, so they are a floor.
DEX liquidity counts pools found in DeFiLlama, DexScreener, Fluid and GeckoTerminal. Pools of Uniswap V4, Balancer and Ekubo that DexScreener does not index are not valued. A coin shows the table only when every chain's pool list is measured complete.
Region figures overlap, because a coin can carry more than one region.

Revisions and versioning

Data revision policy

A later update on the same UTC day replaces the earlier figure.
Charts keep one point per day: the last update of that day.
A page can take up to an hour after an update to show the new figure.
Supply history that cannot be reproduced exactly is not published.

Error correction and restatement

When a published figure is found to be wrong, the corrected figure replaces it. History is restated where it can be reproduced exactly, and the change is noted in the change log.

To report a figure you believe is wrong, use the Talk to us link in the site header.

Methodology review

This methodology is reviewed every quarter. Each review updates the Last reviewed date at the top of this page. A material change gets a new version and a change-log entry with the date it takes effect^[1].

Change log

Changes to this methodology, newest first
Version	Effective	Change
2.1	2026-10-30	The coin page supply chart draws contract supply only, with DeFiLlama's figure before DIA's first day. The circulating line added on 2026-10-19 is removed because circulating supply is stored for only a short part of the window. Circulating supply stays the headline figure. No published figure changes.
2.1	2026-10-29	DEX liquidity totals for the largest coins rise about 7 times, and the coins whose pool list was hidden as a sample now show a table (USDC on Ethereum: $48M in the old 30-pair sample, about $330M in the full list; USDT: $53M, about $380M). Until now the pool list came from DexScreener alone, which returns at most 30 pairs per token in no liquidity order, so for coins with more pools the list missed flagship pools. Pools are now found in DeFiLlama's DEX pools, DexScreener, Fluid and, where a check needs it, GeckoTerminal. The coin's side of each pool is read from DexScreener, the chain or Fluid's deposits (never its borrowed liquidity). A coin shows the table only when every chain's list is measured complete: at least 90% of a 30-pair DexScreener sample's value must also appear in the pool directories. Pools of Uniswap V4, Balancer and Ekubo that DexScreener does not index are left out, and the V4 share is checked against the PoolManager balance. The earlier lists were incomplete samples; coins with more than 30 pools on a chain were hidden rather than shown with them.
2.0	2026-10-19	Restructured as a numbered methodology document with formulas, a glossary, a source inventory, the reconciliation tolerances, limitations, a revision policy and a citation format. Rules the code already applied are now stated: the outlier-leg guard, the price cross-check tolerances, hidden disputed prices, the lending and holder formulas, and reader monitoring.
Edit	2026-10-19	Counts cover stablecoins with measured supply only; a listed coin with none is marked Supply not read or Supply not priced. A chain counts from $100K of supply. Coin-page peer comparisons draw from the counted population (#1510).
Edit	2026-10-19	Peg history leaves out the days a non-dollar coin's whole currency moved together (exchange-rate timing). The home page reports peg figures for dollar, euro and other coins, and the largest closes of the last 30 days (#1508).
Edit	2026-10-19	Holder concentration names its method by source: net of contracts and treasuries, net of treasuries only, or gross (#1506).
Edit	2026-10-19	Contract controls read on Kava, Tron, Aptos, NEAR and TON. A control is marked absent only from a verified source or chain state (#1494).
Edit	2026-10-19	Supply by chain rows add up to the headline: treasuries come off the chain that holds them, bridged supply shows on the chain that holds it, and any remainder is a Not circulating or Not split by chain row (#1493).
Edit	2026-10-19	A dominant home-chain leg is accepted only when the coin's record names the chain, USDS on Ethereum today (#1491). MXNT closes from 2025-04-01 to 2026-01-22, priced from a thin pool, are left out of its peg history (#1490).
Edit	2026-10-19	Circulating supply defined as contract supply less treasury, premint and burn balances. Supply charts draw circulating and contract supply as two series. Copy corrected to 4 updates a day and to the DeFiLlama and CoinGecko price cross-check. Bridged legs, lending share and holder concentration notes added (#1503).
Edit	2026-10-09	Issuer forms, licence tiers and stablecoin layers published. Update times checked against the scheduler.
Edit	2026-10-07	One rule for frozen and destroyed funds, applied to every stablecoin.
Edit	2026-09-30	Wrapped stablecoins listed but left out of every count and total, so each figure uses one population.
Edit	2026-09-25	Coverage counts read from the live registry.
Edit	2026-09-18	Wound-down stablecoins rule rewritten against the code. Issuers classified by the entity that issues the stablecoin.
1.0	2026-09-17	Page first published.

When figures update

Supply, prices and exchange rates update four times a day, in runs that start at 03:00, 09:00, 15:00 and 21:00 UTC. A page can take up to an hour after an update to show the new figure. Charts keep one point per day: the last update of that day.

Inputs, how often they update and when
Input	Frequency	Runs (UTC)	Note
Token contracts	4 a day	03:00, 09:00, 15:00, 21:00	A later update on the same day replaces the earlier figure.
DIA price feed	4 a day	With each supply update	The latest DIA price at each update. A price older than 48 hours is never used. The peg figures on the home page use only readings from the latest update day or the day before.
Exchange rates	4 a day	With each supply update	ECB rates change on business days only.
Block explorers	4 a day	With each supply update	Holder counts and top holders.
Lending markets	4 a day	00:00, 06:00, 12:00, 18:00	Balances, utilisation and supply rates.
DeFiLlama	4 a day	00:30, 06:30, 12:30, 18:30	For stablecoins with no DIA figure.
Price history	Daily	One close per day	One closing price per day from DeFiLlama, or CoinGecko where DeFiLlama has none. Peg history charts draw from it, so they come from DeFiLlama and CoinGecko, not from DIA's price feed.
Contract controls	Weekly	Sundays, 05:00 UTC	Pause, freeze and upgrade functions found in each token contract.
Issuer records	By hand	On review	When DIA reviews a filing, licence or reserve report.

How to cite

The data change with every update, so a citation of the data names the methodology version and the date it was retrieved.

Data set: DIA (2026). Global Stablecoin Map, methodology version 2.1 [Data set]. https://www.diadata.org/map/global-stablecoins/. Retrieved YYYY-MM-DD.
This document: DIA (2026). Global Stablecoin Map: Data Sources & Methodologies, version 2.1. https://www.diadata.org/map/global-stablecoins/data-sources/
In a chart or table: Source: DIA Global Stablecoin Map
A section: DIA (2026), §4.3

References

IOSCO (2013). Principles for Financial Benchmarks: Final Report. International Organization of Securities Commissions. https://www.iosco.org/library/pubdocs/pdf/IOSCOPD415.pdf ↑
Kosse, A., Glowka, M., Mattei, I. and Rice, T. (2023). Will the real stablecoin please stand up? BIS Papers No 141. Bank for International Settlements. https://www.bis.org/publ/bppdf/bispap141.htm ↑
DeFiLlama. Stablecoins, lending and price data. https://defillama.com/ ↑
Coin Metrics. Free Float Supply. Coin Metrics Data Knowledge Base. https://docs.coinmetrics.io/network-data/network-data-overview/supply/free-float-supply ↑
Bank for International Settlements (2025). Stablecoin growth: policy challenges and approaches. BIS Bulletin No 108. https://www.bis.org/publ/bisbull108.htm ↑
European Central Bank. Euro foreign exchange reference rates. https://www.ecb.europa.eu/stats/policy_and_exchange_rates/euro_reference_exchange_rates/html/index.en.html ↑
CoinGecko. Cryptocurrency prices. https://www.coingecko.com/ ↑

Scope and coverage

Objective

Universe and inclusion criteria

A stablecoin is listed when its status is active.
Supply is only ever measured from contract deployments DIA has verified onchain.
3 wrapped stablecoins, such as sDAI, are listed but not counted in any total or count on the map.
A listed stablecoin is counted when it has measured supply: a dollar value above zero, or a first-party total withheld by the coverage rule in section 4.1.2, whose supply exists but is not shown today. Every count on the map uses this population.
1 listed stablecoin has no measured supply yet: it stays listed, marked “Supply not read” when no contract was read or “Supply not priced” when tokens were read but no price exists, and is left out of every count.
An issuer or a peg currency is counted when at least one of its stablecoins is counted.
A chain is counted when listed stablecoins hold $100K+ on it between them, valued as in section 4.1.6. Smaller chains stay listed on the chain page.
A peer comparison on a coin page draws from counted stablecoins only, and each card names the population it ranks against.
Only active stablecoins appear in the ranked tables and in the currency, region and issuer pages.
A stablecoin that has wound down keeps its own page, marked historical, and is excluded from those tables.

Metadata at a glance

Statistical unit: One stablecoin: one token, counted once across all of its chains.
Population: 143 stablecoins with measured supply, from 99 issuers.
Reference area: 72 chains with $100K+ of supply. 25 peg currencies.
Unit of measure: US dollars for value, native token units for supply, basis points for peg deviation, percent for shares.
Reference period: The snapshot taken at each update. Charts keep one point per UTC day: the last update of that day.
Frequency: 4 a day, in runs that start at 03:00, 09:00, 15:00 and 21:00 UTC (section 7.5).
Time coverage: From each stablecoin's first DIA supply reading. Supply charts show DeFiLlama history before that day (section 4.1.5).

Coverage at a glance

143Stablecoins with supply

99Issuers

25Peg currencies

72Chains with $100K+

4 a dayUpdates

48 hoursOldest price used

Definitions and classification

Glossary

Active stablecoin: A stablecoin whose status is active. Only these enter tables and totals; counts take the counted ones. Section 1.2.
Basis point (bp): One hundredth of a percent, 0.01%.
Bridged leg: A leg on one chain that is a claim on tokens already counted on another chain. Section 4.1.1.
Circulating supply: Contract supply less the treasury, premint and burn balances, published only when that list is complete and every balance read. Formula (4.2).
Contract supply: The token total read from the verified contracts that read in a run. Formula (4.1).
Counted stablecoin: A listed stablecoin with measured supply: a dollar value above zero, or a withheld first-party total. Every count on the map uses these. Section 1.2.
Excluded role: A treasury, premint or burn address, whose balance is taken out of circulating supply. Section 4.1.
First-party figure: A figure computed from DIA's own contract reads and DIA's own price. A DeFiLlama figure is labelled as not measured by DIA. Section 3.3.
Headline holder chain: The one chain on which a stablecoin's holders are counted. Section 4.5.
Herfindahl-Hirschman index (HHI): The sum of the squared holder shares in percent, from 0 to 10,000. Formula (4.12).
Holder: An address with a balance above zero on the headline holder chain, frozen addresses included. Section 4.5.
Leg: One stablecoin's supply on one chain.
Lending share: The share of a stablecoin's value supplied to the lending markets DIA reads. Formula (4.9).
Peg currency: The currency a stablecoin is designed to be worth one unit of.
Peg deviation: The distance between a stablecoin's price and one unit of its peg currency, in basis points. Formula (4.6).
Supply not read, Supply not priced: The marks on a listed stablecoin with no measured supply: no contract was read, or tokens were read but no price exists. Such a coin is left out of every count. Section 1.2.
Wrapper: A token that holds another listed stablecoin, such as sDAI. It is listed and left out of every total and count. Section 1.2.

Issuer types and licences

Each stablecoin is listed under the company or protocol that issues it. Each issuer has a form and a licence tier. The rules below run top to bottom, and the first match decides.

When a stablecoin is sold under another company's brand, that company is recorded as its brand owner. The brand does not change the issuer's form. The stablecoin page names both.

Issuer form

What kind of legal thing issues it?

1. Bank
The issuer, or a parent that controls it, holds a deposit-taking banking licence on a regulator's register.
2. Trust company
The issuer holds a trust charter with no deposit-taking: NYDFS limited purpose trust, OCC national trust bank, US state trust company or Canadian provincial trust corporation.
3. Non-bank company
A named legal entity mints and redeems (including through a KYC allowlist), or holds or custodies the reserve off-chain.
4. Protocol
Otherwise: minting and redemption are open to any address meeting onchain conditions, and the reserve sits in smart contracts.

Licence tier

Is it authorised to issue, and where? An issuer takes the highest tier that any of its licences or registrations reaches.

1. Recognised regime
Holds an authorisation covering issuance in a recognised regime (listed below).
2. Licensed elsewhere
Holds an issuance licence from a regulator outside the recognised regimes.
3. Registered elsewhere
Registered as an issuer, or its token or fund registered, with a regulator outside the recognised regimes, and holds no issuance licence.
4. Registration only
Holds only an AML or VASP registration, SRO membership, sandbox or incubation status, a pending application, a securities exemption or a MiCA white paper.
5. None found
We checked the registers and the issuer's own pages and found no licence or registration. Includes protocols with no issuing entity.
6. Not checked
No regime on record: we have not looked for a licence for this issuer yet.

Recognised regimes

A licence puts an issuer in the Recognised regime tier only when it covers issuance and was granted in one of these:

EU and EEA under MiCA (e-money institution or credit institution)
United States (state money transmitter, NYDFS, OCC, GENIUS Act permitted issuer)
United Kingdom (FCA e-money)
Switzerland (FINMA licence, not self-regulatory organisation membership)
Singapore (MAS)
Hong Kong (HKMA)
Japan (JFSA)
UAE (CBUAE, VARA, ADGM, DIFC)
Australia (ASIC AFSL)
Canada (CSA value-referenced crypto asset framework)
The regulator of any jurisdiction whose central bank issues a currency the issuer's coins track

An issuance licence from any other regulator, including El Salvador, Bermuda and the Isle of Man, puts the issuer in Licensed elsewhere.

Stablecoin layers

Stablecoin layers and their values, in evaluation order
Layer	Question	Values, in evaluation order
Backing	What is the coin designed to be backed by?	Own-token / algorithmic. There is no collateral, or the collateral is mainly the issuing protocol's own token. Cash & money-market. Designed to be backed by bank deposits, cash, government bills of a year or less, reverse repo or money-market fund shares (looked through for funds only). Other stablecoins. Designed to be backed by other stablecoins or lending receipts on them, with no look-through. Crypto collateral. Designed to be backed by volatile crypto or staking tokens held as overcollateral, unhedged. Hedged crypto. Designed to be backed by crypto whose price exposure is offset by derivative shorts, spot and hedge counted as one position. Credit & other RWA. Designed to be backed by loans, private credit, corporate paper, government debt over a year, or other real-world assets.
Holder return	Does simply holding it pay you?	Price accrues. An ordinary holder sees a unit's redemption value rise by design (share price, NAV accrual, CPI index). Balance rebases. An ordinary holder's balance grows without any action (rebase, or dividends paid in new tokens). None. Holding pays nothing, including coins whose yield goes only to allowlisted earners or to a separate staked token.
Holder recourse	Who, if anyone, owes you a unit of the peg?	Issuer redemption. A named entity's terms give holders a right to redeem 1:1 at that entity, possibly after onboarding or KYC. Contract only. Redemption runs only through contracts, pools or liquidations; no entity owes the holder a unit of the peg.

Data sources

Source inventory

Primary sources are read by DIA. Reference sources check or complete them. The aggregator, DeFiLlama^[3], stands in only where DIA has no figure. Run times are UTC.

Sources, the metrics they feed, run times and freshness rules
Source	Role	Metrics fed	Runs (UTC)	Freshness rule
On-Chain Contract Reads	Primary	Supply per chain; treasury, premint and burn balances	03:00, 09:00, 15:00, 21:00	A failed chain is a failure, never zero
DIA Price Aggregation	Primary	Price, peg deviation	With each supply update	48 hours; symbol-matched quotes rejected
Fiat Reference Rates	Reference	Dollar rate of non-dollar peg currencies	With each supply update; ECB on business days	No quote, no peg deviation
Peg Cross-Check	Reference	DeFiLlama and CoinGecko current prices	With each supply update	Section 4.3.3
Block Explorers	Reference	Holder counts, top holders	With each supply update	Section 4.5
Lending Markets	Reference	Supplied balances, utilisation, supply rates	00:00, 06:00, 12:00, 18:00	Markets not updated for 3 days left out
DeFiLlama Fallback	Aggregator	Supply with no DIA figure, chain split, history before DIA's first read	00:30, 06:30, 12:30, 18:30	Labelled as not measured by DIA
Price history	Reference	Peg history charts	One close per day	DeFiLlama, or CoinGecko where DeFiLlama has none
Contract controls	Primary	Pause, freeze and upgrade functions	Sundays, 05:00	Section 4.6
Issuer Research	Manual	Identity, jurisdiction, licence, reserves, transparency links	On review of a filing, licence or reserve report	Each row carries its documents

Source descriptions

On-Chain Contract Reads

Primary

Refresh4 times a day

DIA Price Aggregation

Primary

Prices come from DIA's own aggregation across trading venues. A quotation matched by symbol instead of by the specific contract is rejected. A price older than 48 hours is never used.

Refresh4 times a day, prices older than 48 hours not used

Peg Cross-Check

Reference

Before a peg deviation is shown, DIA's price is compared with the current price from DeFiLlama and from CoinGecko. The rule is in section 4.3.3.

Refresh4 times a day

Fiat Reference Rates

Reference

Refresh4 times a day, ECB rates change on business days

Block Explorers

Reference

Holder counts and top holders come from block explorers such as Blockscout and Etherscan. The headline holder count is taken on one chain per stablecoin.

Refresh4 times a day

Lending Markets

Reference

Balances deposited in the lending markets DIA reads for its vault map. The stablecoin pages show how much of each stablecoin sits in them. A market not updated for three days is left out.

Refresh4 times a day

Issuer Research

Manual

RefreshManual review

DeFiLlama Fallback

Aggregator

Coins with no DIA figure yet fall back to DeFiLlama for the supply figure and are not labelled as measured by DIA.

Refresh4 times a day

Source hierarchy

DIA's own contract reads and DIA's own price come first. A figure built from them is a first-party figure.
Where DIA has no supply figure or no price for a stablecoin, DeFiLlama's circulating figure stands in, and the figure is labelled as not measured by DIA.
For a stablecoin no market source quotes, a curated exchange rate, or for a staked version the vault's share price times the price of the coin it holds, can value the supply. Such a price carries no peg reading.
Sources are never mixed inside one figure without a label. A total DIA withholds is not replaced with DeFiLlama's figure.

Calculation methods

Circulating supply

S_{contract} = \sum_{c \in C} (T_{c} - K_{c})

(4.1)

S_{circ} = S_{contract} - \sum_{a \in E} b_{a}

(4.2)

where E is the curated list of treasury, premint and burn addresses on the chains in C, and b_a is the balance of address a. An address on a chain outside C is not subtracted.

Withheld when

the coin's exclusion list is not curated, cited and marked complete;
any balance in E fails to read;
the result is below zero or above S_contract.

In each case the headline is contract supply. Where circulating supply publishes, issuer-held supply is S_contract minus S_circ.

Bridged legs

Withholding rules

Rules that withhold or label a supply total
Rule	Applies when	Effect
Unmapped bridged legs	Chains hold bridged claims on tokens counted on another chain, and the bridge is not mapped	Total and DIA price withheld; legs still shown
Mint and burn cycling	The contract mints and burns on a cycle, so one read is a point-in-time balance	Total and DIA price withheld
Dominant unissued supply	Almost all of the onchain supply is unissued	Total and DIA price withheld
Concentrated premint	The holder scan finds 10 or fewer holders and 95% or more in one address	Total and DIA price withheld
Coverage regression	A chain read in at least 3 of the last 5 published snapshots is missing from this run	Total withheld until the chain reads again or leaves the window
Outlier leg	Section 4.1.3	Total withheld; the outlier leg is dropped
Unissued overhang	The issuer states that some onchain supply is not in circulation	Published, labelled unissued
Premint concentration	Curated: almost the whole supply sits in a handful of wallets and has barely moved	Published, labelled concentrated

Outlier-leg guard

Frozen and destroyed funds

Value, fallback and history

V = S \times P

(4.3)

where S is the published supply (circulating where it publishes, contract supply otherwise) and P is the published price in US dollars.

Where no first-party supply or no price exists, V is DeFiLlama's circulating dollar figure, labelled as not measured by DIA. A total withheld under section 4.1.2 has no value.
The supply chart uses DeFiLlama history only for days before DIA's first supply reading. A day is never drawn from both sources. A step of more than 3% at the junction is marked on the chart. When the two series differ by more than 5 times at the junction, the DeFiLlama history is dropped.
The supply chart draws DIA’s contract supply, with DeFiLlama’s figure before DIA’s first day. Circulating supply is stored for only a short part of the window, so it is the headline figure and is not charted.

Per-chain split

The Supply by chain rows on a coin page add up to the headline.

Each issuer treasury comes off the chain that holds it.
Bridged supply shows on the chain that holds it and comes off its source chain. A bridged leg that cannot be placed, because its source chain is not counted or has nothing left, lists its read and stays out of the sum.
Where the rows carry more than the headline, as with a bridge escrow or a treasury with no per-chain split, the difference is one Not circulating row, and shares are of the chain rows. Where the headline is above the rows, the difference is one Not split by chain row.
No shares are shown when the total is withheld or is not a first-party figure.

Market aggregates

V_{total} = \sum_{i \in A} V_{i}

(4.4)

where A is the set of active stablecoins, wrappers out, that have a value V_i from (4.3).

w_{k} = \frac{\sum_{i \in A_{k}} V_{i}}{V_{total}}

(4.5)

where A_k is the subset of A pegged to currency k.

Supply by chain is the sum of each coin's chain values from section 4.1.6. The home page shows the 10 largest chains.
crvUSD enters every total at its debt-based circulating supply (section 4.1).
A stablecoin counts once in each region it is tagged with, so region figures overlap and do not add up to the total.

Peg deviation

Peg deviation measures a price against par, one unit of the peg currency^[5].

d = round (10^{4} \times (\frac{P_{USD}}{R_{peg}} - 1))

(4.6)

Eligibility

Freshness

A price older than 48 hours is never used. The peg figures on the home page use only readings from the latest update day or the day before.

Cross-check

Before a peg deviation is shown, DIA's reading is compared with the readings from the current DeFiLlama and CoinGecko^[7] prices.

| d_{DIA} - d_{r} | \leq \max (τ, 0.15 \times \max (| d_{DIA} |, | d_{r} |))

(4.7)

where d_r is the deviation from (4.6) at reference r, DeFiLlama or CoinGecko, and τ is 50 bp, or 150 bp for yen coins.

The figure shows when (4.7) holds for at least one reference that has a price. It is withheld when DIA disagrees with every reference that has a price.
A DeFiLlama reading more than 1,000 bp from both DIA and CoinGecko is treated as a failed source and ignored.
With no reference price, DIA's reading stands as unchecked. A DIA price within 1 bp of a daily fallback exchange rate is then treated as that rate, not a market price, and withheld.
A disputed DIA price is hidden as well when the previous run published no peg reading for the coin, or when the price is identical to the previous run's price.

Home-page figures

Today's peg figures count stablecoins with a current reading and at least $1M of supply, wrappers out. The headline share of supply within 50 bp of $1 and the distribution in bands at 10, 50 and 100 bp cover dollar coins.
The same readings are grouped by peg currency: dollar, euro, and every other currency together. Each group reports how many coins sit within 10, 50 and 100 bp, and the median deviation. A non-dollar reading is DIA's dollar price divided by the day's exchange rate, so it carries that rate's timing.
A 30-day panel lists, for dollar coins, the closes furthest from $1 over the last 30 days and how many coins closed more than 100 bp away. It reads the stored daily closes from DeFiLlama and CoinGecko, so a coin with no reading today can still appear.
A deviation of more than 500 bp carries one sentence with the date of the reading and what DeFiLlama shows for the coin.

Peg stability over time

σ = \sqrt{\frac{1}{n} \sum_{t \in W} {(d_{t} - \bar{d})}^{2}}

(4.8)

where W is the 90 calendar days ending on the last close (30 days for the short figure), n the number of closes in it and d̄ their mean. With fewer than 2 closes there is no figure.

Largest move: the close furthest from par in the last 12 months, with its date and sign.
Peer ranking: where the coin's 90-day volatility sits among the active, non-accruing coins of its peg currency with a 90-day figure, the coin included. It is shown only when at least 3 coins have one. The peer chart and table cover dollar coins only.
Exchange-rate timing: a non-dollar close is a dollar price divided by an exchange rate taken at a different moment, so a currency move between the two prints the same false move on every coin of that currency. A day is left out of a non-dollar coin's largest move when all of these hold: the coin is at least 50 bp from par; at least 3 other coins of the same peg currency closed that day; a strict majority of them sit on the same side of par within max(25 bp, 35% of the coin's deviation) of it; and their median is no more than 200 bp from par. The page says how many days were left out.
Left out as provider errors: a single close more than 2,000 bp from both neighbouring days when those two are within 500 bp of each other; filled closes of exactly 1.0000 before 2025-10-01; MXNT closes from 2025-04-01 to 2026-01-22, priced from a thin trading pool; and a short list of dated prints. Stored data is not changed, and the chart notes each exclusion.

The share of a coin's onchain supply that sits in lending markets DIA reads.

L_{i} = \min (1, \frac{M_{i}}{S_{i} \times P_{i}})

(4.9)

where M_i is the dollar value supplied to the counted markets, S_i the onchain supply and P_i DIA's price, or a fallback market price between 0.01 and 10 times the peg.

In: markets that report a balance and an update within 3 days.
Out: any market holding less than $1,000 of the coin; abandoned markets whose balance grows only from accrued interest (utilisation of 99.9% or more with a supply rate of 50% or more); DEX liquidity, CDP collateral and isolated collateral.
Rebasing coins show no share. The figure is a floor, since chains and protocols DIA does not read are not counted.

L = \frac{\sum_{i} \min (M_{i} - M_{i}^{self}, V_{i})}{\sum_{i} V_{i}}

(4.10)

the home-page share across stablecoins, where M_i^self is the issuer's own supply of its stablecoin to those markets and V_i is from (4.3). Wrappers and rebasing coins are out of both sums.

Holders and concentration

A holder is an address with a balance above zero, frozen addresses included.
Holders are counted on one chain per stablecoin: the largest chain an explorer can read, which must exceed the default chain, Ethereum where the coin has it, by more than 5%. The default chain keeps a tie. When that chain fails to read, the next largest chain holding at least 10% of the readable supply is used.

s_{N} = 100 \times \frac{\sum_{j = 1}^{N} b_{(j)}}{S_{h}}

(4.11)

where b_(j) is the j-th largest balance, N is 1, 5, 10, 50 or 100, and S_h is the coin's verified supply on that chain.

HHI = \sum_{j} {(100 \times \frac{b_{j}}{S_{h}})}^{2}

(4.12)

over every holder j, on a scale of 0 to 10,000.

A top-N figure is published only when the holder list read reaches N addresses or covers every holder, and is left blank otherwise. HHI is published only when the read covers every holder.
No share is published when the balances read add up to more than the chain's supply.
What is left out depends on what the source returns, and the coin page names the method under its table:
- Net of contracts and treasuries (Blockscout, and Ethplorer, which keeps Blockscout's shares): the issuer's treasury wallets and burn addresses leave the holder list and the supply; addresses named as protocol or bridge contracts leave the holder list but keep their supply in the denominator.
- Net of treasuries (Tronscan, NodeReal and DIA's own replay of transfers): treasuries and burn addresses are netted the same way. These sources name no contracts, so contracts count as holders.
- Gross (Routescan and the Solana cluster): these sources return no addresses, so nothing is netted and shares are of the whole supply on the chain.
Exchange wallets always stay in.
The map publishes these measures and attaches no labels to them.

Contract controls

Data quality and validation

Failure handling

A chain that fails to read is a failure, never a zero. A coin with no successful read reports no supply.
A withheld figure is never replaced by an estimate.
One unreadable treasury, premint or burn balance withholds circulating supply (section 4.1).
An unread holder count is not a zero count. A failed holder scan never withholds supply on its own.
A lending read that returns nothing shows no lending figure, not 0%.

Price cross-check

Every peg reading is checked against two independent prices before it is shown. The rule is in section 4.3.3.

Independent reconciliation

Each day, every published total is compared with an independent directory. For each pair the relative gap is

δ = \frac{| x_{ours} - x_{ref} |}{\max (x_{ours}, x_{ref})}

(5.1)

where x_ours is DIA's figure and x_ref the reference figure, both in US dollars.

Each pair takes the first severity that applies:

missing ours: our figure is below the floor and the reference is at or above the missing threshold;
missing reference: the reference is below the floor and ours is at or above the missing threshold;
ok: the absolute gap is below the floor;
large: δ is at or above the large percentage and the absolute gap at or above the large amount;
minor: δ is at or above the minor percentage;
ok otherwise.

Reconciliation tolerances by lane
Lane	Reference	Minor	Large	Floor	Large amount	Missing threshold
Supply	DeFiLlama stablecoins	5%	20%	$1M	$25M	$10M
Lending	DeFiLlama lending, Morpho	15%	40%	$0.5M	$10M	$5M
Supply per chain	DeFiLlama stablecoins, per chain	5%	10%	$1M	$50M	$50M

Further checks run in their own units:

Reconciliation checks with their own thresholds
Check	Compares	Minor	Large	Scope
Peg today	DIA peg deviation against CoinGecko's	25 bp	75 bp	A missing reading is raised for the 30 largest coins only
Peg volatility	90-day volatility of the daily deviation, recomputed	10 bp	25 bp	The 20 largest coins
Unchecked share	Share of dollar coins with no published peg reading	Not used	Above 20%	Coins with a snapshot in the last 7 days
Overhang	Published supply more than 25% above DeFiLlama's token count, with 70% or more in the top 5 holders	Not used	Both conditions hold	Excess of $1M or more
Backing classes	Cash and money-market total against DeFiLlama's fiat-backed total	A coin sits in only one of the two	More than 2% apart	Coins matched in both
Holder counts	Stored holder count against a second explorer read	5%	20% and 1,000 holders	Every 7 days, the 20 largest coins on EVM chains
Yields	Shown pool rates against DIA's onchain lending rate	1.0 points and 25%	3.0 points and 50%	Pools shown in the last 14 days with $5M or more deposited

Reader freshness monitoring

Silent: a reader with no successful write for 24 hours.
Dropped: a run that writes less than half the median of the last 7 days, where there are at least 4 earlier runs and the median is 5 rows or more.
Never written: a registered reader with 3 or more attempts and no rows. A lending or supply reader with no rows at all alerts on its first attempt.
Stale markets: a live reader whose markets holding $1M or more have not been written for 3 days. Morpho markets are left out, because the Morpho reader drops abandoned and dust markets on purpose.

Manual research controls

Limitations

The figures have known limits^[1]:

Holder counts and concentration cover one chain per stablecoin.
HHI and the top 50 and top 100 shares need a complete holder read, so many coins show none.
Peg history charts come from DeFiLlama and CoinGecko closing prices, not DIA's price feed.
Figures from the DeFiLlama fallback are not first-party.
ECB reference rates move on business days only.
A non-dollar peg reading divides a dollar price by an exchange rate taken at a different moment, so it carries the timing gap between the two (section 4.3.5).
A currency that no source quotes shows no peg figure.
Issuer research is updated by hand and can lag a new filing.
Lending figures cover only the markets DIA reads, so they are a floor.
DEX liquidity counts pools found in DeFiLlama, DexScreener, Fluid and GeckoTerminal. Pools of Uniswap V4, Balancer and Ekubo that DexScreener does not index are not valued. A coin shows the table only when every chain's pool list is measured complete.
Region figures overlap, because a coin can carry more than one region.

Revisions and versioning

Data revision policy

A later update on the same UTC day replaces the earlier figure.
Charts keep one point per day: the last update of that day.
A page can take up to an hour after an update to show the new figure.
Supply history that cannot be reproduced exactly is not published.

Error correction and restatement

When a published figure is found to be wrong, the corrected figure replaces it. History is restated where it can be reproduced exactly, and the change is noted in the change log.

To report a figure you believe is wrong, use the Talk to us link in the site header.

Methodology review

Change log

Changes to this methodology, newest first
Version	Effective	Change
2.1	2026-10-30	The coin page supply chart draws contract supply only, with DeFiLlama's figure before DIA's first day. The circulating line added on 2026-10-19 is removed because circulating supply is stored for only a short part of the window. Circulating supply stays the headline figure. No published figure changes.
2.1	2026-10-29	DEX liquidity totals for the largest coins rise about 7 times, and the coins whose pool list was hidden as a sample now show a table (USDC on Ethereum: $48M in the old 30-pair sample, about $330M in the full list; USDT: $53M, about $380M). Until now the pool list came from DexScreener alone, which returns at most 30 pairs per token in no liquidity order, so for coins with more pools the list missed flagship pools. Pools are now found in DeFiLlama's DEX pools, DexScreener, Fluid and, where a check needs it, GeckoTerminal. The coin's side of each pool is read from DexScreener, the chain or Fluid's deposits (never its borrowed liquidity). A coin shows the table only when every chain's list is measured complete: at least 90% of a 30-pair DexScreener sample's value must also appear in the pool directories. Pools of Uniswap V4, Balancer and Ekubo that DexScreener does not index are left out, and the V4 share is checked against the PoolManager balance. The earlier lists were incomplete samples; coins with more than 30 pools on a chain were hidden rather than shown with them.
2.0	2026-10-19	Restructured as a numbered methodology document with formulas, a glossary, a source inventory, the reconciliation tolerances, limitations, a revision policy and a citation format. Rules the code already applied are now stated: the outlier-leg guard, the price cross-check tolerances, hidden disputed prices, the lending and holder formulas, and reader monitoring.
Edit	2026-10-19	Counts cover stablecoins with measured supply only; a listed coin with none is marked Supply not read or Supply not priced. A chain counts from $100K of supply. Coin-page peer comparisons draw from the counted population (#1510).
Edit	2026-10-19	Peg history leaves out the days a non-dollar coin's whole currency moved together (exchange-rate timing). The home page reports peg figures for dollar, euro and other coins, and the largest closes of the last 30 days (#1508).
Edit	2026-10-19	Holder concentration names its method by source: net of contracts and treasuries, net of treasuries only, or gross (#1506).
Edit	2026-10-19	Contract controls read on Kava, Tron, Aptos, NEAR and TON. A control is marked absent only from a verified source or chain state (#1494).
Edit	2026-10-19	Supply by chain rows add up to the headline: treasuries come off the chain that holds them, bridged supply shows on the chain that holds it, and any remainder is a Not circulating or Not split by chain row (#1493).
Edit	2026-10-19	A dominant home-chain leg is accepted only when the coin's record names the chain, USDS on Ethereum today (#1491). MXNT closes from 2025-04-01 to 2026-01-22, priced from a thin pool, are left out of its peg history (#1490).
Edit	2026-10-19	Circulating supply defined as contract supply less treasury, premint and burn balances. Supply charts draw circulating and contract supply as two series. Copy corrected to 4 updates a day and to the DeFiLlama and CoinGecko price cross-check. Bridged legs, lending share and holder concentration notes added (#1503).
Edit	2026-10-09	Issuer forms, licence tiers and stablecoin layers published. Update times checked against the scheduler.
Edit	2026-10-07	One rule for frozen and destroyed funds, applied to every stablecoin.
Edit	2026-09-30	Wrapped stablecoins listed but left out of every count and total, so each figure uses one population.
Edit	2026-09-25	Coverage counts read from the live registry.
Edit	2026-09-18	Wound-down stablecoins rule rewritten against the code. Issuers classified by the entity that issues the stablecoin.
1.0	2026-09-17	Page first published.

When figures update

Inputs, how often they update and when
Input	Frequency	Runs (UTC)	Note
Token contracts	4 a day	03:00, 09:00, 15:00, 21:00	A later update on the same day replaces the earlier figure.
DIA price feed	4 a day	With each supply update	The latest DIA price at each update. A price older than 48 hours is never used. The peg figures on the home page use only readings from the latest update day or the day before.
Exchange rates	4 a day	With each supply update	ECB rates change on business days only.
Block explorers	4 a day	With each supply update	Holder counts and top holders.
Lending markets	4 a day	00:00, 06:00, 12:00, 18:00	Balances, utilisation and supply rates.
DeFiLlama	4 a day	00:30, 06:30, 12:30, 18:30	For stablecoins with no DIA figure.
Price history	Daily	One close per day	One closing price per day from DeFiLlama, or CoinGecko where DeFiLlama has none. Peg history charts draw from it, so they come from DeFiLlama and CoinGecko, not from DIA's price feed.
Contract controls	Weekly	Sundays, 05:00 UTC	Pause, freeze and upgrade functions found in each token contract.
Issuer records	By hand	On review	When DIA reviews a filing, licence or reserve report.

How to cite

The data change with every update, so a citation of the data names the methodology version and the date it was retrieved.

Data set: DIA (2026). Global Stablecoin Map, methodology version 2.1 [Data set]. https://www.diadata.org/map/global-stablecoins/. Retrieved YYYY-MM-DD.
This document: DIA (2026). Global Stablecoin Map: Data Sources & Methodologies, version 2.1. https://www.diadata.org/map/global-stablecoins/data-sources/
In a chart or table: Source: DIA Global Stablecoin Map
A section: DIA (2026), §4.3

References

IOSCO (2013). Principles for Financial Benchmarks: Final Report. International Organization of Securities Commissions. https://www.iosco.org/library/pubdocs/pdf/IOSCOPD415.pdf ↑
Kosse, A., Glowka, M., Mattei, I. and Rice, T. (2023). Will the real stablecoin please stand up? BIS Papers No 141. Bank for International Settlements. https://www.bis.org/publ/bppdf/bispap141.htm ↑
DeFiLlama. Stablecoins, lending and price data. https://defillama.com/ ↑
Coin Metrics. Free Float Supply. Coin Metrics Data Knowledge Base. https://docs.coinmetrics.io/network-data/network-data-overview/supply/free-float-supply ↑
Bank for International Settlements (2025). Stablecoin growth: policy challenges and approaches. BIS Bulletin No 108. https://www.bis.org/publ/bisbull108.htm ↑
European Central Bank. Euro foreign exchange reference rates. https://www.ecb.europa.eu/stats/policy_and_exchange_rates/euro_reference_exchange_rates/html/index.en.html ↑
CoinGecko. Cryptocurrency prices. https://www.coingecko.com/ ↑

1Scope and coverage#

1.1Objective#

1.2Universe and inclusion criteria#

1.3Metadata at a glance#

2Definitions and classification#

2.1Glossary#

2.2Issuer types and licences#

Issuer form

Licence tier

Recognised regimes

2.3Stablecoin layers#

3Data sources#

3.1Source inventory#

3.2Source descriptions#

On-Chain Contract Reads

DIA Price Aggregation

Peg Cross-Check

Fiat Reference Rates

Block Explorers

Lending Markets

Issuer Research

DeFiLlama Fallback

3.3Source hierarchy#

4Calculation methods#

4.1Circulating supply#

4.1.1Bridged legs#

4.1.2Withholding rules#

4.1.3Outlier-leg guard#

4.1.4Frozen and destroyed funds#

4.1.5Value, fallback and history#

4.1.6Per-chain split#

4.2Market aggregates#

4.3Peg deviation#

4.3.1Eligibility#

4.3.2Freshness#

4.3.3Cross-check#

4.3.4Home-page figures#

4.3.5Peg stability over time#

4.4Lending share and utilisation#

4.5Holders and concentration#

4.6Contract controls#

5Data quality and validation#

5.1Failure handling#

5.2Price cross-check#

5.3Independent reconciliation#

5.4Reader freshness monitoring#

5.5Manual research controls#

6Limitations#

7Revisions and versioning#

7.1Data revision policy#

7.2Error correction and restatement#

7.3Methodology review#

7.4Change log#

7.5When figures update#

8How to cite#

9References#

Building with stablecoins?

1Scope and coverage#

1.1Objective#

1.2Universe and inclusion criteria#

1.3Metadata at a glance#

2Definitions and classification#

2.1Glossary#

2.2Issuer types and licences#

Issuer form

Licence tier

Recognised regimes

2.3Stablecoin layers#

3Data sources#

3.1Source inventory#

3.2Source descriptions#

On-Chain Contract Reads

DIA Price Aggregation

Peg Cross-Check

Fiat Reference Rates

Block Explorers

Lending Markets

Issuer Research

DeFiLlama Fallback

3.3Source hierarchy#

Scope and coverage

Objective

Universe and inclusion criteria

Metadata at a glance

Definitions and classification

Glossary

Issuer types and licences

Stablecoin layers

Data sources

Source inventory

Source descriptions

Source hierarchy

Calculation methods

Circulating supply

Bridged legs

Withholding rules

Outlier-leg guard

Frozen and destroyed funds

Value, fallback and history

Per-chain split

Market aggregates

Peg deviation

Eligibility

Freshness

Cross-check

Home-page figures

Peg stability over time

Lending share and utilisation

Holders and concentration

Contract controls

Data quality and validation

Failure handling

Price cross-check

Independent reconciliation

Reader freshness monitoring

Manual research controls

Limitations

Revisions and versioning

Data revision policy

Error correction and restatement

Methodology review

Change log

When figures update

How to cite

References

Scope and coverage

Objective

Universe and inclusion criteria

Metadata at a glance

Definitions and classification

Glossary

Issuer types and licences

Stablecoin layers

Data sources

Source inventory

Source descriptions

Source hierarchy

Calculation methods

Circulating supply

Bridged legs

Withholding rules

Outlier-leg guard

Frozen and destroyed funds

Value, fallback and history

Per-chain split

Market aggregates

Peg deviation

Eligibility

Freshness

Cross-check

Home-page figures

Peg stability over time

Lending share and utilisation

Holders and concentration

Contract controls

Data quality and validation

Failure handling

Price cross-check

Independent reconciliation

Reader freshness monitoring