Nvidia posts $96.2B quarter; UC Merced and Fresno State weigh what it means
Nvidia’s blowout quarter and higher guidance signal tight demand for AI chips. UC Merced and Fresno State projects that rely on GPUs will watch prices and lead times.
Nvidia posts $96.2B quarter; UC Merced and Fresno State weigh what it means
Key Takeaways
- Nvidia reported Q2 fiscal 2027 revenue of $96.2 billion, up 106% year over year.
- The company guided next quarter’s revenue to $108 billion, plus or minus 2%.
- Data Center revenue hit $89.0 billion, and Edge Computing reached $7.2 billion.
- UC Merced’s Pinnacles cluster includes Nvidia A100, L40S, and H200 GPU nodes.
- Fresno State’s new F3i Supercomputing Center launched to support Valley ag-tech work.
Nvidia posted $96.2 billion in quarterly revenue on Wednesday. UC Merced and Fresno State run research gear that depends on the same chips everyone else wants. When Nvidia beats forecasts and raises guidance, it touches GPU prices, wait times and the pace of campus projects here.
What Nvidia reported
The chipmaker’s quarter ended July 26 and cleared Wall Street’s bar by a wide margin. Nvidia said Data Center sales reached $89.0 billion, with its Edge Computing line at $7.2 billion. The outlook called for $108 billion in revenue for the August to October period, a figure that keeps pressure on supply even if the mix shifts between cloud customers and everyone else. Shares rose in after-hours trading.
Why it matters in the Valley
Campus labs and applied research groups don’t buy at hyperscale, they buy on grants and calendars. That means delivery times and bid prices can speed up or slow down a semester’s worth of work in Fresno County and Merced County. UC Merced’s high‑performance setup, the Pinnacles cluster, lists 20 GPU nodes that include A100 accelerators for training, newer L40S boards for inference and visualization, and H200 NVL units with high‑bandwidth memory. Those are Nvidia parts. Which is the point.
Farm problems need compute too.
Fresno State just stood up the F3i Supercomputing Center with San Diego Supercomputer Center and ag‑tech nonprofit F3 Innovate to give growers, students and startups shared access to heavy compute and AI help. The pitch is simple: don’t build a private machine if a regional center can get you cycles, storage and expertise faster. If Nvidia’s guidance holds, the market impact lands on procurement desks from Shields Avenue to Lake Road.
What local campuses are building
UC Merced’s documentation shows GPU nodes tied into fast InfiniBand networking and a support team that trains researchers to schedule jobs and move data efficiently. Fresno State’s center is set up to run applied projects across the food and agriculture supply chain and to train students who will take those skills to co‑ops, processors and irrigation companies in the Valley. Fresno State’s Computer Science labs also note Nvidia hardware on the desktop side, a reminder that not every model needs a pod to be useful.
A half‑empty Dr Pepper sat on a lab bench during a recent walk‑through.
The near term question is pace. If cloud vendors keep taking the first slice of every shipment, smaller buyers will time their purchases around grant windows and batch availability. If supply broadens, UC Merced and Fresno State can add nodes, push new classes and open more seats in training programs without waiting on backorders that slip past finals. Either way, the hum out of a new server room in Fresno tells you the work has already started.
Central Valley AI is produced by the CVAI Newsdesk team and developed by Kaweah Tech, a regional firm that builds, deploys, and integrates AI solutions for businesses across California's Central Valley.