Bitmine Immersion Technologies picked up another 9,926 ether (ETH) during the week ending Aug. 16. The purchase pushed its treasury to roughly 5.82 million ETH, putting the company within reach of a 5% share of total Ethereum supply. The position, meanwhile, sits more than $8.4 billion underwater.
The latest buy cost about $19 million, according to Decrypt. It extends an unbroken streak that dates to June 2025, when Bitmine launched its ETH treasury strategy and started accumulating without pause, Cointelegraph reported.
The firm now holds approximately 4.8% of Ethereum’s circulating supply. Decrypt describes that as 96% of the way to Bitmine’s stated long-term target, a goal the company refers to as its “Alchemy of 5%” owning 5% of total ETH supply. The two figures use different denominators, circulating versus total supply. Both point the same direction. Bitmine is close.
At an ETH reference price of $1,893, the portfolio carried a value of roughly $11 billion. Ether traded little changed on Monday, sitting just above $1,900.
Those paper gains are not gains. Bitmine is sitting on more than $8.4 billion in unrealized losses, according to industry data from DropsTab. The loss ratio runs about 43%. That is a mark-to-market figure against the firm’s average cost, and it has done nothing to slow the accumulation.
Tom Lee, the Fundstrat founder who chairs Bitmine, has framed staking income as the counterweight. Bitmine is staking more than 5 million ETH, worth roughly $9.6 billion at current prices. Based on a seven-day staking yield of 2.61%, Lee projects annualized staking rewards of approximately $287 million.
That projection depends on assumptions holding. The $287 million figure requires yield to stay at 2.61% over a full year and the staked position to remain intact. Staking yields on Ethereum are variable. Network participation and protocol issuance set them, not Bitmine. If more ETH gets staked network-wide, the per-validator yield falls. That is how the mechanism works.
The buying pattern raises a structural question that staking revenue does not answer. A single corporate entity controlling close to 5% of Ethereum’s supply is a concentration event. A meaningful slice of the network’s active supply ends up locked in one treasury, one staking validator set, one governance posture. Whether that concentration translates into governance influence depends on decisions Bitmine has not detailed publicly.
What is on the record is the pace. Bitmine has bought every week since June. The losses have grown alongside the holdings because the average cost basis sits well above the current spot price. DropsTab’s data puts the unrealized loss at $8.4 billion, climbing whenever ETH slips. The firm has not disclosed its exact cost basis, nor the individual purchase prices that make up the average.
Last week’s $19 million buy is small against an $11 billion position. It is a maintenance purchase, enough to keep the streak alive and nudge holdings toward the 5% line. Not enough to move the cost basis. At 96% of the target, the remaining gap is roughly 130,000 ETH, depending on which supply denominator applies.
Bitmine has not stated a timeline for reaching 5%. Lee’s $287 million staking projection is the only public revenue forecast. The unrealized loss is the only public mark on the cost side.
Holders are divided on whether the model works, as they tend to be. The staking income is real and recurring as long as the ETH stays staked. The $8.4 billion hole is also real. It deepens every time ether gives back ground. Both can be true at once.
