MANTRA DAO Initiates Survey Vote to Solicit Feedback on Team Token Burn Execution

Pre-market key news for US stocks is as follows: 1. Deutsche Bank cuts PayPal’s price target from $60 to $55; raises Meta’s price target from $750 to $820. 2. Strategy plans to launch daily dividend mechanisms for STRF, STRC, STRK and STRD. 3. Citigroup remains bullish on US stocks, maintains an overweight stance on US equities, and will increase investment exposure during market pullbacks. 4. Following the release of US durable goods data, the yield on the 30-year US Treasury note rose 1.85 basis points to 5.481%, while the 10-year US Treasury note yield climbed 2.36 basis points to 5.186%. 5. Goldman Sachs says AI-related capital expenditure is set to surge to $1.7 trillion, requiring $1.42 trillion in revenue over three years to achieve a 15% return. 6. US stock index futures rose as oil prices and US Treasury yields pulled back from recent sharp gains, with markets awaiting key inflation expectation data. 7. Most of the US "Magnificent Seven" stocks advanced in pre-market trading; Akamai jumped 20% after signing a $11.6 billion deal with Anthropic. 8. According to Japan’s Kyodo News: Honda is considering investing 400 billion yen to build a hybrid vehicle factory in Ohio. 9. Microsoft launched a new Copilot with Home, Code and Autopilot features, integrating the full functionality of Word, Excel and PowerPoint into Copilot via "Office in Copilot". 10. Tesla has increased production of its Optimus humanoid robot by roughly 10 times in recent months.

Beating AI News Flash: Per market sources, Nvidia has pledged a $1 billion investment in AI computing power firm Nscale, with the funds expected to be in place mid-November. On September 19 earlier this year, AI data center operator Nscale filed an IPO application with the U.S. SEC, planning to list in New York and raise up to $3 billion under the proposed ticker symbol NSCL. The company spun off from its cryptocurrency mining business in early 2024 and has since transitioned to become an AI computing infrastructure service provider. According to the prospectus, as of August 2026, Nscale owns and controls data centers with a power capacity exceeding 10GW, with total contract value of roughly $103 billion, including a $45 billion lease commitment from Anthropic this August. The company’s clients and partners include Nvidia, Microsoft, and other firms.

Arkham disclosed in a post that Hyperliquid address "MK4" currently holds a NEAR long position worth around $29.7 million, with an entry price of approximately $2.35, using 10x leverage, and generating an unrealized profit of roughly $15.96 million. The blockchain analytics firm added that the address has accumulated about $58 million in total profits on Hyperliquid over the past three years, and is suspected to belong to X account @mk4_lul.

Bernstein said in a report that a sharp rise in memory chip costs could pressure Apple's iPhone margins in its December quarter, raising the risk of underperforming expectations. The firm has cut its earnings per share (EPS) forecast for Apple from $3.00 to $2.87, below the market consensus estimate of $2.92. Despite expecting near-term margin pressure, Bernstein maintained its "Outperform" rating on Apple and $370 price target, noting that Apple remains a key entry point for consumer AI.

According to Beating AI Express, citing a report from The Information, Microsoft has launched a redesigned "super app" version of Copilot, along with a new product competing with Muse. Microsoft CEO Satya Nadella stated that the redesign aims to "make Copilot the new operating system for work scenarios." The new Copilot comes with updated Home, Code, and Autopilot functions, and integrates the full capabilities of Word, Excel, and PowerPoint into the tool via "Office in Copilot." Earlier, on September 9, Meta released its personal agent Muse. The biggest difference between Muse and OpenClaw and Hermes is that it does not require self-deployment, nor does it involve the hassle of managing models, servers, and skills. Meta directly provides an independent cloud environment, allowing users to access it through apps, web pages, or WhatsApp.

Beating AI Express News: Max For AI stated that domestic large models are currently entering an "awkward period", with core challenges focusing on three areas: training computing power, inference resources, and commercial pricing. As domestic vendors continue to expand model scales, sizes of 5 trillion, 8 trillion, and even 10 trillion parameters are being discussed, but the growth rate of training computing power may struggle to keep pace with this expansion, making computing power a practical constraint. Alibaba previously noted at its Cloud Summit that it may train models with 5T to 10T parameters, while vendors including Kimi and GLM are also advancing large-scale model development. On the inference side, completing large model training does not equate to being able to stably serve a large user base long-term. Max For AI pointed out that some large models run fast at launch, but may later face speed limits, queuing, and tightened quotas, reflecting that inference computing remains a key bottleneck. As model scales grow, each additional user quota translates to higher actual inference costs. Additionally, rising model sizes and inference costs may force vendors to adjust pricing, as one of the key competitive advantages of domestic large models has been open-source availability and cost-effectiveness. The firm added that in the Agent era, as single tasks consume more tokens, price sensitivity will further increase, requiring domestic vendors to rebalance model performance, inference costs, and commercial pricing.