Thursday's US stock market saw NVIDIA as the second-highest traded stock, closing up 1.80% with $30.5 billion in volume. The chip giant has agreed to acquire AI startup Hugging Face for approximately $13 billion, with the deal including an equity retention plan worth up to $1 billion to keep Hugging Face employees who join NVIDIA on board. The company stated it will maintain the open nature of the Hugging Face platform and continue the startup's existing operational approach. This acquisition represents CEO Jensen Huang's largest move yet to expand AI accessibility and broaden the customer base. The deal will give NVIDIA control over a key platform where developers showcase and share AI models. Additionally, NVIDIA announced that NVIDIA RTX Spark Windows PCs will hit the market in October 2026.

Tesla Motors ranked third, surging 5.42% with $23.6 billion in trading volume. On September 3 local time, the company is set to officially unveil its autonomous taxi, the Cybercab. Morgan Stanley highlighted in its latest research note that this launch carries far more substantive significance than any previous product event—the scaled expansion of a driverless Robotaxi fleet will directly determine the stock's trajectory for the rest of the year. If Tesla Motors successfully deploys 25 to 50 Cybercabs for paid ride-hailing services in Texas within days to weeks after the event, the stock is expected to react positively. Conversely, if the event only showcases limited supervised retail demonstrations without substantial driverless fleet deployment, the stock is likely to face downward pressure afterward.

Micron Technology finished fourth, edging up 0.22% with $22.77 billion in volume. Reports indicate the company is researching a high-durability NAND flash memory module designed to be positioned closer to GPUs. The product concept, tentatively called "near-GPU NAND," aims to give GPUs direct access to NAND storage pools of several hundred gigabytes, with the storage modules placed either inside the GPU package or on the PCB board, operating similarly to existing high-bandwidth memory solutions like HBM and GDDR7. Compared to traditional TLC or QLC NAND, this chip offers lower storage density but focuses on enhanced I/O speed and bandwidth.

Broadcom took the fifth spot, dropping 2.74% with $21.1 billion in volume. The company's fourth fiscal quarter revenue is projected at approximately $34.8 billion, up 93% year-over-year but below the analyst consensus of $35.05 billion. Within this, AI semiconductor revenue guidance stands at $21.7 billion, a 236% year-over-year increase, only slightly above the market expectation of $21.33 billion.

SpaceX ranked seventh, climbing 6.42% with $18.02 billion in trading volume. The company has applied to the US Federal Communications Commission (FCC) for permission to carry and operate Starlink terminals during the fourteenth test flight of Starship. According to the submitted documents, the mission plans to conduct a true orbital flight—something the Starship program has never attempted before. Oppenheimer analysts, led by Timothy Horan, stated in a Wednesday client note that SpaceX is poised to become a winner in the short-term computing power shortage. Horan gave SpaceX a buy rating and raised the target price from $250 to $280, implying roughly 100% upside from current levels.

SanDisk Corp. came in eighth, inching up 0.10% with $13.37 billion in volume. Microsoft placed ninth, gaining 2.68% with $12.31 billion traded. The software giant announced it will disclose Azure sales on a quarterly basis starting fiscal year 2027, and will merge the Intelligent Cloud and Productivity & Business Processes segments into a new "Agents & Infrastructure" division. Azure sales for the fiscal year ending in June exceeded $100 billion, up from $75 billion the prior year, representing approximately 30% of total company revenue.

Apple ranked tenth, rising 1.00% with $12.20 billion in volume. TrendForce estimates the price increase for the new iPhone will be controlled between 10% and 20%, with the iPhone Fold series expected to start between $2,099 and $2,299. Morgan Stanley, in a Wednesday email report, said Apple's upcoming first foldable iPhone could contribute approximately $14 billion to December quarter revenue. The brokerage projects shipments of around 6.5 million units in the December quarter, generating roughly $14 billion in revenue—about 16% of total iPhone sales for that period.

Meta Platforms, Inc. came in eleventh, advancing 3.01% with $12.06 billion in volume. The company released Muse Spark 1.3, its most powerful AI model to date. Chief AI Officer Alexandr Wang described this as the company's "biggest leap" in model performance, particularly in coding and agent tasks, bringing Meta Platforms, Inc. closer to top competitors like OpenAI and Anthropic.

Dell Technologies Inc. took twelfth place, jumping 4.91% with $10.34 billion traded. In the second quarter of fiscal year 2027, ending July 31, the company posted revenue of $46.97 billion, up 58% year-over-year and above the analyst average of $44.92 billion. Net income for the quarter was $4.13 billion, or $6.34 per share, more than tripling from $1.16 billion and $1.70 per share a year earlier. Adjusted earnings per share (excluding stock-based compensation) came in at $7.04, far exceeding the analyst forecast of $4.92. Looking ahead, Dell Technologies Inc. provided strong third-quarter guidance: adjusted EPS of $6.50 on revenue of $49 billion, implying approximately 81% year-over-year revenue growth.

Snowflake ranked fourteenth, surging 16.55% with $8.04 billion in volume. The company reported better-than-expected second quarter results, with revenue growing 35% year-over-year to $1.55 billion and core product revenue up 37% to $1.49 billion. The company raised its full-year guidance, now expecting product revenue to grow 36% to approximately $6.07 billion.

Alphabet Class A shares finished sixteenth, adding 1.59% with $7.08 billion in volume. Google DeepMind and Google Research announced WeatherNext 3 today, touted as the most advanced and accurate global weather model currently available. The model learns directly from real-time observations, enabling more timely and localized forecasts for weather events affecting people.