10 #Bitcoin ETFs inflows 2,586 $BTC(+$247.92M) and 9 #Ethereum ETFs outflows 20,884 $ETH(-$70.69M) yesterday.

According to Farside's monitoring, U.S. spot Bitcoin ETFs recorded a net inflow of $730.8 million yesterday, with BlackRock's IBIT attracting a net inflow of $454 million and ARKB seeing a net inflow of $137.7 million.

According to Farside’s monitoring, U.S. Ethereum spot ETFs posted a net inflow of $141.4 million yesterday, with BlackRock’s ETHA attracting $72.1 million in net inflows and Fidelity’s FETH bringing in $65.1 million.

Beating AI Express (Dongcha) News: Alibaba Cloud announced in a post that its Qianwen Office has reached a one-month milestone since launch, with over 30 million office users, more than half of whom are enterprise users. Qianwen Office has maintained high-frequency iterations over the past month, rolling out a total of 120 version updates (averaging 4 iterations per day) and adding over 1,000 features. Its international version is also now live. Qianwen Office has open-sourced its context base, the enterprise context infrastructure MyContext, which converts massive heterogeneous work data into contexts directly callable by Agents. Just weeks after its open-source release, MyContext has earned over 3,000 stars on GitHub. Integrated with deep connectivity to instant messaging (IM) tools like DingTalk, Qianwen Office can convert enterprise information scattered across business processes—including messages, documents, approvals, emails, and knowledge bases—into organizational contexts that Agents can understand and utilize. A simple command in DingTalk enables Agents to query customer data, check inventory, and provide recommendations across CRM and ERP systems. Together with the Qianwen large language model (LLM) team, Qianwen Office has launched the office-exclusive version Qwen3.8-Flash, which has been specially optimized for multi-step planning, tool selection, and context compression. It further enhances throughput efficiency via inference optimization and a customized Harness architecture. In real office scenarios, the generation speed of individual tasks has improved by roughly 100%, with token consumption reduced by an average of 75%.

According to GMGN data, the crypto-stock meme project MEME on Robinhood Chain saw its market capitalization briefly surge past $11 million to a new high, currently trading at $11.2 million. The token has rallied over 600 times in 24 hours, with a trading volume of $7.8 million. MEME is paired with tokenized U.S. stock AMC Entertainment (the U.S. theater chain), under the ticker AMC. Additionally, another crypto-stock meme project CINEMA on Robinhood is also paired with AMC, making the two tokens liquidity competitors. BlockBeats Note: Crypto-stock meme (Stock Meme) is an emerging concept that combines traditional meme coins with tokenized U.S. stocks: instead of pairing with USDT or ETH, meme coins are directly matched with on-chain U.S. stock tokens (such as NVDA, TSLA, AAPL, etc.). This model retains the high volatility and community-driven speculative traits of meme coins while leveraging the popularity and narrative of real stocks. Part of transaction fees often flows back to the community treasury to accumulate corresponding U.S. stock tokens, forming a dual-driven model of "sentiment speculation + real asset anchoring". Note: Price fluctuations are significant, invest with caution.

Robinhood CEO Vlad Tenev stated: "Robinhood originally launched to serve a new generation of first-time investors. Today, this client base is accumulating wealth, and their financial needs have grown accordingly. Just weeks after rolling out Trust Accounts, client deposits have surpassed $150 million, with an average account size of roughly $500,000. We aim to accompany and serve them throughout their entire financial lifecycle."

Beating AI News reports that Thinking Machines Lab, founded by former OpenAI CTO Mira Murati, is in talks for a new financing round. The company plans to raise at least $10 billion, with a pre-money valuation of no less than $40 billion. Existing investor Accel is in discussions to lead the round, while Nvidia is also considering participating in the deal, which has not yet been finalized. This valuation is significantly higher than the previous round: when Thinking Machines raised $2 billion last year, its pre-money valuation stood at $10 billion, with a post-money valuation of $12 billion. Calculated on the same pre-money basis, the new round’s valuation is at least four times that of over a year ago. Notably, the company had sought a valuation of more than $50 billion at the end of last year, a target it has since lowered. Thinking Machines has also begun generating substantial revenue, with annualized income of at least several hundred million dollars. Revenue primarily comes from Tinker, a platform where customers pay to access its computing power and tools to fine-tune AI models on their own data. The firm’s first self-developed open-weight model, Inkling, has also been integrated into Tinker. This year, Thinking Machines signed a long-term computing power partnership with Nvidia, planning to deploy at least 1GW of the Vera Rubin system, with Nvidia making an additional investment. The new financing round will be used for model training, server leasing, and talent recruitment.