ONCHAINBENCHMARKThe measurement standard for tokenized capital
IndicesDataRatingsMethodologyResearchReportsAboutLog inOpen Terminal ↗
Reports/Stablecoins/$1.08B
Stablecoins

The Day $1.75 Billion of USDT ‘Left’ Ethereum — and $670 Million of It Never Left Tether’s Own Wallet

The market is calling it a stablecoin supply war: USDC gaining, USDT shrinking. On August 10, the largest dollar token's raw supply on Ethereum dropped $1.75 billion in a single intra-day step. But most of that was Tether moving its own inventory: measured at circulating supply, about $1 billion relocated to Tron and roughly $670 million simply came out of Tether's own Ethereum treasury wallet. Net circulating change across the two chains: essentially zero. That is a chain swap — mechanics, not demand — visible onchain the moment it happens.

Key findings
  • →On August 10, 2026, USDT's raw supply on Ethereum fell $1.75 billion in a single intra-day step while its raw supply on Tron rose about $1.0 billion. Measured at circulating supply — net of Tether's own onchain treasury — about $1 billion relocated from Ethereum to Tron and the net change across the two chains was essentially zero. It is the onchain fingerprint of a chain swap, Tether's own mechanism for moving inventory between networks, not demand leaving the token. see data →
  • →Roughly $670 million of the widely-quoted $1.75 billion Ethereum drop never left circulation: it came out of Tether's own Ethereum treasury wallet. Raw supply counts tokens the issuer has minted but not issued; netting them out turns an apparent large 'burn' into a clean ~$1 billion cross-chain swap.
  • →Aggregate supply figures compress issuer mechanics, treasury inventory and genuine demand into one number. On a token spanning a dozen-plus networks, a portion of any month's change in USDT supply is inventory relocating between chains, or between circulation and the issuer's own treasury, rather than dollars entering or leaving the token. Separating the two requires measuring each chain on the same, treasury-aware definition, daily.
OnChain Benchmark Research
Published · Aug 11, 2026, 05:07 PM UTCData current · Aug 11, 2026

Figures are as of the publication date. Current figures are on the record.

Revised October 2, 2026: figures corrected.

For most of this year the stablecoin story has been told as a scoreboard: Circle's USDC gaining, Tether's USDT slipping, total supply pressing toward record highs. Read chain by chain, though, a large part of what looks like “USDT shrinking” turns out not to be demand at all. On August 10, the largest dollar token's raw supply on Ethereum fell $1.75 billion — in a single intra-day step — while its raw supply on Tron rose about a billion in the same window. And roughly $670 million of that Ethereum drop never left Tether's own wallet.

That is not a token bleeding out. It is a chain swap: the routine mechanics by which the biggest stablecoin moves its own inventory from one network to another. Tether has documented the process — it reduces supply on the network it is leaving and authorizes an equivalent amount on the network it is expanding into, and it distinguishes tokens it has authorized and holds in its own treasury from tokens actually issued into circulation. Raw supply counts both. Net the treasury out and the event sharpens: about $1.0 billion of circulating USDT relocated to Tron, and about $670 million of the Ethereum figure was Tether's own inventory leaving its Ethereum treasury wallet, not dollars leaving holders.

Circulating USDT off Ethereum, Aug 10$1.08B$87.87B → $86.79B, net of Tether's treasury
Circulating USDT onto Tron, same window$1.01B$89.88B → $90.89B; the matching leg of the swap
Of the $1.75B raw Ethereum drop: treasury, not circulation~$670Mcame out of Tether's own Ethereum wallet

An event, not a trend

The distinction matters because the two look identical in an aggregate supply chart. USDT's Ethereum supply sat unchanged through the morning of August 10; by mid-day it had stepped down and it has held there since. Tron moved in lockstep the other way. This was not a slow migration over weeks — it was one discrete step, and a rare one: across the 90 days to that date, only two single-day moves in USDT's Ethereum supply exceeded half a billion dollars. The chart below shows the raw Ethereum line, the circulating Ethereum line beneath it, and Tron rising to meet the difference.

Fig 1 — USDT on Ethereum (raw vs circulating) and Tron (circulating) · daily · Aug 6–11, 2026

The raw Ethereum drop is bigger than the circulating one — the gap is Tether's own treasury

USDT on Ethereum by raw supply (light grey) and circulating supply, net of Tether's own treasury (dark grey), and USDT on Tron by circulating supply (green), in billions. All three hold flat, then step in a single intra-day move on August 10. The raw Ethereum line falls $1.75B; the circulating line falls only about $1.08B; Tron's circulating line rises about $1.01B. The gap between the two Ethereum lines is Tether's own treasury inventory, not tokens in circulation.

ONCHAIN
BENCHMARK
USDT SUPPLY ($B) DATE $86 $88 $90 $92 $93 Aug 6 Aug 7 Aug 8 Aug 9 Aug 10 Aug 11 $90 $87 $91 ETH raw ETH circulating Tron circulating
Source: OnChain Benchmark per-chain supply, net of Tether's onchain treasury (same-block pair) · onchainbenchmark.comas of Aug 11, 2026
See USDT coverage →

On the circulating measure the two legs very nearly net. Ethereum lost about $1.08 billion of circulating USDT; Tron gained about $1.01 billion; the net change in circulating USDT across the two chains was essentially zero. What made the Ethereum figure look larger was Tether's treasury: at the moment of the swap the raw Ethereum supply fell $1.75 billion, but $1.75 billion also came out of Tether's own Ethereum wallet in the same step, so circulation did not move at that instant. The circulating drop registered a few hours later, as the treasury refilled while raw supply held — which is why raw supply and circulating supply tell different stories about the same day. We measure both, at the same block; we do not read intent into it.

What a supply number leaves out

Return to the scoreboard with this in view. A headline that reads “USDT supply on Ethereum fell $1.75 billion” is technically true and mostly beside the point: about a billion of it relocated to Tron and about two-thirds of a billion never left Tether's own wallet. Supply figures compress issuer mechanics, treasury inventory and genuine demand into a single number, and on a token that spans more than a dozen networks, the mechanics are often the larger part. The same treasury adjustment is why our companion measure of chain-level dollar concentration reads USDT at circulating rather than raw supply. The only way to separate demand from mechanics is to measure each chain on the same definition, every day, and net out what the issuer is holding for itself.

None of this settles whether USDC is genuinely gaining share on USDT — it may well be. It does mean the raw supply deltas the debate leans on are noisier than they look: a chunk of any month's “USDT shrank” is inventory moving between Ethereum and Tron, or between circulation and Tether's own treasury, rather than dollars leaving the token. The next time a single day's supply number swings by a billion, the useful question is not “how much shrank” but “which chain, whose wallet, and where did it go” — and that is a question a treasury-aware, cross-chain measurement can answer while it is still happening.

The eventPer-chain supply, gross and circulating (ground truth)Re-derived, same-block pair

On 2026-08-10, USDT raw supply on Ethereum stepped from $92.06B to $90.31B (−$1.75B) in a single intra-day interval and held; USDT raw supply on Tron rose from $90.28B to $91.28B (+~$1.0B) in the same window. Netting Tether's own onchain treasury at the same block: Ethereum circulating fell from $87.87B to $86.79B (−$1.08B), Tron circulating rose from $89.88B to $90.89B (+$1.01B). Figures are raw onchain totalSupply minus the treasury balance at the same block, 6 decimals, re-derived.

Verified Sep 26, 2026 · onchainbenchmark.com/instruments/usdt
Circulating vs rawWhy the $1.75B is bigger than the swapMeasured

At the 08:29–08:30 step, $1.75B left both Ethereum's raw supply and Tether's Ethereum treasury wallet ($4.18B → $2.43B), so circulation did not move at that instant; the ~$1.08B circulating drop registered over the following hours as the treasury refilled while raw supply held. About $670M of the raw Ethereum drop was therefore Tether's own treasury inventory leaving its wallet, not tokens leaving circulation. Net circulating change across the two chains: about −$0.08B, essentially flat.

OnChain Benchmark methodology
ScopeWhat it measuresDisclosed

This measures Tether's own documented chain-swap mechanism on the two chains that moved (Ethereum and Tron), each read on a single treasury wallet at the same block as supply. It is not a claim about total USDT across every chain, nor about who transacted or why. 'Authorized' inventory held in treasury versus 'issued' tokens in circulation follows Tether's own issuance primer.

Tether Issuance Primer · tether.io
MethodologySupply is the raw onchain totalSupply of each USDT deployment. Circulating supply nets out Tether's own treasury balance — tokens it has minted but not issued — read from its treasury wallet at the same block as the supply figure, so there is no timing skew. On Ethereum and Tron that treasury is a single wallet with one reading per block. The August 10 step is identified from the intra-day read sequence: flat before and after, one discrete move between. “Chain swap” and the authorized-versus-issued distinction follow Tether's own issuance primer. Revised Sep 26, 2026: figures corrected. Full methodology: onchainbenchmark.com/methodology.
Questions & answers
Did USDT lose $1.75 billion in demand on August 10, 2026?
No. USDT's raw supply on Ethereum fell $1.75 billion, but about $1.0 billion of circulating USDT reappeared on Tron in the same window, and roughly $670 million of the Ethereum drop was Tether's own treasury inventory rather than tokens leaving circulation. Measured at circulating supply, the net change across the two chains was essentially zero. That is a chain swap — Tether moving inventory between networks — not demand leaving.
What is the difference between raw supply and circulating supply here?
Raw onchain supply is every token that exists on a chain, including tokens Tether has minted into its own treasury but not yet issued. Circulating supply nets that treasury out. During a chain swap the two diverge sharply: on August 10 Ethereum's raw supply fell $1.75 billion, but about $670 million of that was Tether's own treasury inventory, so circulating supply fell only about $1.08 billion. OnChain Benchmark reads USDT at circulating supply, from the treasury balance at the same block as the supply figure.
What is a stablecoin chain swap?
It is the mechanism an issuer uses to move its token's supply from one blockchain to another: it reduces (burns) supply on the network it is leaving and authorizes an equivalent amount on the network it is expanding into. Tether documents the process, including a distinction between 'authorized' inventory held in its own treasury for future issuance and 'issued' tokens in active circulation. Onchain, a swap looks like a large same-day drop on one network paired with a rise on another — and, once the treasury is netted out, a much smaller change in what is actually circulating.
Does this mean USDC is not gaining on USDT?
It does not settle that question either way. USDC may well be gaining share. The point is narrower: the raw supply deltas the debate relies on are noisier than they look, because a portion of any month's change in USDT supply is inventory relocating between chains, or between circulation and Tether's own treasury, rather than dollars entering or leaving the token. Separating the two requires measuring each chain on the same, treasury-aware definition, daily.
Key facts
Circulating USDT off Ethereum (Aug 10)$1.08B
Circulating USDT onto Tron (same window)$1.01B
Net circulating change, two chains≈ flat
Of the $1.75B raw drop: treasury, not circulation~$670M
BasisCirculating (net of issuer treasury) ↗
More reports
Nov 10Stablecoins
One Depositor Pulled 15% of a US$4.5 Billion Stablecoin Savings Pool in a Day
Nov 8Tokenized Treasuries
BlackRock's BUIDL Shrank 35% in 90 Days. A Holder of a Quarter of the Fund Left Entirely.
Oct 31Stablecoins
An Estimated US$6.3 Billion Stolen, US$2.8 Billion Frozen: Why the Binance Hacker Still Holds US$1.2 Billion in Ether
Oct 30Stablecoins
Four Hours, 20 Freezes, One Attacker: Inside the Race to Freeze the Binance Hack, Second by Second
Oct 27Tokenized Equities
Two in Three Addresses Holding Solana's Largest Tokenized-Stock Line Hold Less Than a Dollar
Understand the ratings
What is an RWA quality rating?
How a tokenized asset is rated
The methodology
The full four-dimension rubric
Go deeper
The current ratings behind this report are in the terminal.
Full quality rating, evidence trail, and per-deployment supply history.
See USDT coverage ↗
OnChain Benchmark — 03

More from Reports

All reports →
StablecoinsNov 10
One Depositor Pulled 15% of a US$4.5 Billion Stablecoin Savings Pool in a Day
Tokenized TreasuriesNov 8
BlackRock's BUIDL Shrank 35% in 90 Days. A Holder of a Quarter of the Fund Left Entirely.
StablecoinsOct 31
An Estimated US$6.3 Billion Stolen, US$2.8 Billion Frozen: Why the Binance Hacker Still Holds US$1.2 Billion in Ether
StablecoinsOct 30
Four Hours, 20 Freezes, One Attacker: Inside the Race to Freeze the Binance Hack, Second by Second
Tokenized EquitiesOct 27
Two in Three Addresses Holding Solana's Largest Tokenized-Stock Line Hold Less Than a Dollar
Tokenized TreasuriesOct 20
Spiko's Euro T-Bill Fund Is 30% Smaller. The Accounts Behind Most of Its Arbitrum Redemptions Now Hold Spiko's Amundi Fund.
OnChain Benchmark
HomeRatingsIndicesFlowsMethodologyReportsAboutEditorial policy
Independent · 2026

Subscribe to the Weekly Benchmark — what moved onchain this week, and what it means.

One email a week. Unsubscribe anytime.

Important disclosures

Nothing on this site or in the OnChain Benchmark service is investment, financial, legal, tax, or accounting advice, an offer or solicitation to buy or sell any security, token, or other instrument, or a recommendation, endorsement, or rating of suitability regarding any instrument, issuer, or transaction. Quality ratings, dimension scores, and index values are independent analytical opinions based on the data available to us at the time of computation; they are not statements of fact, guarantees of quality or safety, and not endorsements. Outputs are produced from public sources and third-party data and infrastructure outside our control — including RPC and node providers, indexers, oracle feeds, issuer attestations, auditor reports, and public regulatory filings — any of which may be delayed, unavailable, incomplete, incorrect, manipulated, or revised after the fact. Scores are computed periodically and may not reflect events occurring after the most recent computation. OnChain Benchmark does not warrant that any score, index, or evidence item is accurate, complete, reliable, current, or suitable for any purpose. You are solely responsible for any decision you make using the service, you should not rely on it as the sole basis for any investment, trading, custody, or counterparty decision, and you should conduct your own independent diligence and consult your own qualified professional advisors. To the maximum extent permitted by law, you assume all risk arising from your use of, or reliance on, the service. See our Terms of Service for the complete terms governing your use of the service.