BREAKING: $435 million has been liquidated from the cryptocurrency market in the past 12 hours.

Bitget has launched a US stock earnings prediction contest running from July 21 to mid-to-late August. During the event, users can earn prediction tickets by completing stock spot (rToken) or stock contract trading tasks, which grant access to predicting the rise or fall of US stock earnings. Prediction results are determined by comparing the closing price of the first trading day after earnings release with that of the trading day prior to the earnings announcement. Different prediction sessions will be rolled out successively based on US stock earnings release timelines, with the first batch covering assets including Tesla, Google, and Intel. Each session has a total prize pool of 10,000 USDT, split between winning and losing participants at a 70% to 30% ratio, and the maximum individual payout per session is 1,000 USDT. For more details, please refer to Bitget’s official platform.

According to Bitget market data, spot silver has risen more than 2% intraday, currently trading at $57.56 per ounce. Spot gold has rallied $8 in the short term, hitting $4037 per ounce.

According to Lookonchain's monitoring, whale address 0x4cee, which had been dormant for three months, has become active again, spending 20 million USDC to purchase 10,501 ETH at an average price of $1,905. Additionally, a newly created wallet address has withdrawn 12,800 ETH from Binance over the past 24 hours, valued at approximately $24.47 million, and has staked all of the ETH.

Base co-founder Jesse Pollak stated in a post that Robinhood Chain’s introduction of tokenized stocks in the EVM environment is a move it got right, while expressing regret that Base had lagged behind in this space. Pollak added that Base is set to partner with Coinbase to address this gap, with plans to launch a tokenized stock product backed by 1:1 stock assets. He noted that compared with Robinhood’s derivative-based model, this solution will offer better scalability in terms of trust, capital efficiency, and institutional adoption.

According to Hyperinsight monitoring, CRCL on Hyperliquid is currently trading at $66.14, up 9.15% from yesterday, marking the largest gain on Trade.xyz’s traditional markets, with an intraday trading volume of approximately $48.25 million. A whale address starting with 0x352 continuously built long positions in CRCL from the evening of July 16 to July 19, accumulating 53,700 units, making it the most profitable long whale in this rally. It currently holds CRCL long positions worth around $3.554 million with 10x leverage, at an average entry price of $61.016, generating an unrealized profit of ~$275,000 (an 84% return), with a liquidation price of $51.05 and no pending CRCL orders at present. Meanwhile, the same address is bottom-fishing SPCX at an average price of $127, currently posting a $680,000 loss, with a position size of $11.67 million. All prior transactions of this address were centered on the 2026 World Cup predictions under the HIP-4 ecosystem, with a cumulative profit of ~$37,000 from the World Cup champion market, mainly from bets on Argentina and Spain winning the title. On the news front, Trump has agreed to the ethics provisions in the CLARITY Act, clearing a key barrier for the crypto legislation to advance to a Senate vote. Circle received final approval from the Office of the Comptroller of the Currency (OCC) on July 10 to establish a national trust bank; BNY Mellon and Standard Chartered have recently expanded USDC custody, minting and redemption services respectively.

After the launch of Kimi K3, the U.S. semiconductor sector declined as markets feared a "DeepSeek Moment 2.0". However, UBS, Nomura, BofA Securities, and Citigroup all agree that the model does not reduce demand for computing power, but may instead drive further expansion of demand for AI infrastructure. Kimi K3 features 2.8 trillion parameters, a 1 million-token context window, and supports always-on inference, native multimodality, and Mixture of Experts (MoE) architecture. Institutions note that its scale and long context will increase KV cache occupancy, boosting demand for HBM, server DDR5, enterprise-grade solid-state drives (SSDs), cloud infrastructure, and high-speed interconnects. Citigroup dubs this trend "another Jevons Paradox", explaining that improvements in model efficiency and cost could lead to more applications and token consumption. UBS points out that open-source models typically rely more on memory and storage due to their longer context windows; Citigroup estimates that large-scale deployment of K3 may require supernodes composed of more than 64 GPUs. Nomura, meanwhile, argues that global competition in large language models (LLMs) will push leading labs and hyperscale cloud platforms to continue investing, benefiting AI infrastructure segments including TSMC, NVIDIA, storage vendors, optical module suppliers, and data center operators. That said, BofA Securities warns that if the pace of model efficiency gains continues to outstrip workload growth while actual usage fails to expand in lockstep, AI infrastructure construction could still see a notable pullback.