A Hyperunit whale who once held over $10 billion in BTC lost $250 million due to going long on ETH on Hyperliquid.
PANews reported on February 11th that, according to Arkham, a Hyperunit whale lost $250 million on the Hyperliquid platform due to its long ETH position, the largest ever held on the platform. It is said to have held Bitcoin for nearly 10 years and is apparently linked to Garrett Jin. Arkham speculates that the whale is Chinese and has held Bitcoin since 2017-2018. In the first six months of 2018, it purchased over 100,000 Bitcoins. It held over 90% of its position for seven years until August 2025—after Tom Lee began making large-scale purchases of Ethereum—when its strategy changed.
At its peak, this whale held $11.14 billion worth of Bitcoin. He transferred $4.49 billion worth of BTC to Hyperunit for sale (39,738 BTC, now worth $2.74 billion). He then received $4.07 billion worth of ETH from Hyperunit (886,371 ETH, now worth $1.78 billion) and staked $3.94 billion worth of ETH (856,372 ETH, now worth $1.72 billion) on the Beacon Chain. This whale connected with Garrett Jin through a $40,000 on-chain transfer and with Jack Yi's Trend Research by depositing millions of dollars into the same Binance deposit address. These three parties likely know each other.
Currently, the Hyperunit whale has lost approximately $3.5 billion on its AAVE leveraged ETH position and on-chain BTC/ETH holdings, and $2.22 billion on its staked Ethereum. Earlier this month, due to leveraged liquidation, it further lost $250 million on Hyperliquid, and is now essentially breaking even on that platform. Since its profit and loss (PnL) reached its all-time high, its total realized and unrealized losses are approximately $5.7 billion. The Hyperunit whale maintains a custom entity on the Arkham platform containing 100 different Bitcoin and Ethereum addresses.
You May Also Like
House Judiciary Rejects Vote To Subpoena Banks CEOs For Epstein Case
Propel to Report Q4 and Full Year 2025 Financial Results and Announces Dividend Increase
