Bitmine Immersion Technologies will stop buying ethereum once its holdings reach 5 percent of circulating supply, chairman Tom Lee said Wednesday at the Token2049 conference in Singapore. The company sits about 100,000 ETH short of that ceiling, roughly six to seven weeks of purchases at its recent pace.
The statement puts an end date on what has been one of ether’s most reliable sources of demand. Bitmine has bought ETH every single week since launching its treasury strategy in June 2025. Last week alone it added about $41 million in ether, lifting its stash to 6,016,414 coins, or roughly 4.9 percent of all supply.
“We thought this would take five years,” Lee said on stage. “It took us a little over a year. More importantly, we did this all in the middle of a bear market. Now we’re going to stop.”
A hard cap, not a soft target
Lee called the 5 percent threshold a “hard cap” and framed it as a matter of shareholder value, not market ambition. The company brands the campaign the “Alchemy of 5%” and has treated the ratio as the point of the exercise: a treasury position large enough to matter, but not one that keeps growing without limit.
The cap also shapes what Bitmine does with coins it earns without buying. The company had 5.07 million ETH staked as of its latest disclosure, which makes it one of the largest single staking entities on the network. Lee has previously suggested Bitmine could sell ether earned through staking rather than allow its share of supply to drift above 5 percent, which would add recurring sell flow to a market already short a large buyer.
At current staking yields of roughly 3 percent a year, a 5 percent stake on Bitmine’s stack would generate close to 300,000 ETH annually in rewards, about $770 million at Wednesday’s price. Whether that income gets sold, restaked or redeployed is now a market question of genuine size.
“We only need to get another 100,000 ETH to get to 5%,” Lee said on stage at Token2049, according to Coin Bureau’s coverage of the conference.
What the buying did to the market
For more than 15 months, Bitmine’s orders landed in the market every week regardless of price. Through the run-up that took ether past $4,900 in August 2025 and through the slide that followed, the company kept accumulating. Tracking data puts its average cost well above current levels, and the paper position shows roughly $4.5 billion in unrealized losses with ether trading near $2,570.
The buying was not ceremonial. Bitmine became a NYSE-listed company whose entire balance sheet story is ether, and its disclosures moved the token the way MicroStrategy’s bitcoin filings once moved that coin. Analysts who followed weekly treasury updates counted Bitmine’s purchases as a floor under price, and funds pitched allocations on the strength of the weekly bid.
| Bitmine ether position, latest disclosure | Figure |
|---|---|
| Total holdings | 6,016,414 ETH |
| Share of circulating supply | About 4.9% |
| ETH staked | 5.07 million |
| Net purchases last week | About $41 million |
| Unrealized losses on the stack | Roughly $4.5 billion |
| Date the 5% milestone moved close | October 4, 2026 |
That steady bid is exactly what the market is about to lose. Traders noticed immediately. Ether dropped 5.5 percent during Asian hours on Wednesday, falling harder than bitcoin and most major tokens, with the slide coinciding with Lee’s remarks from Singapore. Grimstone-style leverage liquidations followed once again: more than $400 million of long positions were flushed in a single hour on Wednesday by one market estimate, reviving memories of the October 10 cascade that erased leveraged books across every major exchange.
ETF flows heading the wrong way
The timing is not kind. US spot ether ETFs posted their sixth straight day of net outflows on Tuesday, bleeding about $202 million, the largest daily withdrawal since mid-September. Every dollar of it left BlackRock’s ETHA fund. Bitcoin funds, by contrast, took in $118.9 million the same day, which made the ether side look worse by comparison.
Weekly flows across the wider crypto ETF market still landed positive, at $2.39 billion, enough to push 2026 net inflows back above zero. But the composition matters for ether: a natural buyer stepping away just as six sessions of ETF redemptions stack up gives sellers one more reason to press the price.
Not every read on the market is sour. Citi raised its 12-month bitcoin forecast from $82,000 to $113,000 this week and set an ether target of $3,028, while Bitwise argued that faster agency rulemaking did more for crypto in 2026 than the stalled CLARITY Act would have. Grayscale’s research head Krista Lynch said this week that SEC listing standards now cover about 15 tokens, letting issuers design selective ETFs beyond bitcoin and ether. The bulls have not left the building, but the mechanical demand that carried ether through 2025 is thinning out, week by week.
Why 5 percent became the target
The logic behind the ratio is straightforward. A treasury holding 5 percent of a proof-of-stake asset controls a meaningful share of validation power and earns a sizeable stream of staking rewards, while remaining below the levels that would draw antitrust-style scrutiny or force exchanges and index providers to make awkward decisions about concentration.
There is precedent for caps as salesmanship. MicroStrategy’s bitcoin accumulation has always been framed as continuous rather than capped, which is one reason its own share premium became the story. Bitmine’s leadership chose the opposite framing, announcing a fixed endpoint before the plan started, then hitting the milestone 15 percent ahead of the original five-year schedule. Lee put it bluntly on stage: the company did in a year what it planned to do in five.
The endpoint also protects the equity story. A treasury company that keeps buying in a falling market converts every weekly purchase into more unrealized loss on the books. The same tracking data showing $4.5 billion underwater is the kind of number that short sellers build slides around. Stopping near the top of the league table, with the largest ether position in existence, hands investors a narrative of discipline instead of obsession.
What happens next
Three things to watch. First, the exact date: at roughly $41 million a week, the final purchase lands in early-to-mid November, unless ether rallies and pushes the market-cap share over 5 percent sooner. Second, whether the company follows through on selling staking rewards, which would replace a weekly buy with a weekly sell and turn the largest ETH treasury into a known source of regular supply. Third, whether ETF outflows continue; if the funds stop bleeding and price stabilizes, the loss of Bitmine’s bid will be digestible, and if they do not, holders will be absorbing close to 100 percent of the supply pressure themselves.
There is a fourth, quieter risk. Treasury company premiums, the gap between share price and net asset value, have compressed across the sector, and token holders increasingly treat each company’s ETH stack as overhang rather than support. Once the largest ether treasury announces it is done, that reframing gets harder to argue against. Bitmine’s weekly buys were never an obligation. They were a choice, and the choice ends in six or seven weeks.