Get the latest crypto news, updates, and reports by subscribing to our free newsletter.
Giấy phép số 4978/GP-TTĐT do Sở Thông tin và Truyền thông Hà Nội cấp ngày 14 tháng 10 năm 2019 / Giấy phép SĐ, BS GP ICP số 2107/GP-TTĐT do Sở TTTT Hà Nội cấp ngày 13/7/2022.
© 2026 Index.vn
Hanoi has submitted an important proposal to the Pacific Construction Group (China) regarding the Văn Cao – Hòa Lạc metro line, a project with total investment of 72.3 trillion dong. On the afternoon of April 9, a Memorandum of Understanding on cooperation was signed between the Hanoi Department of Construction…
GPU chips from Nvidia—specialized processors priced at more than $30,000—have become a key backbone for training and inference across many leading AI models. Yet even inside Nvidia, access to additional compute resources is limited, according to comments made at the HumanX conference in San Francisco.
Bryan Catanzaro, who leads Nvidia’s applied deep learning research, said internal teams still have to “scramble” to secure more compute resources, Fortune reported. He attributed the situation to broader industry constraints, saying the entire sector must operate under limited supply.
Catanzaro also pointed to the company’s leadership decisions around allocation, noting that CEO Jensen Huang often has to decline requests for more GPUs because they have already sold out.
Catanzaro is leading Nemotron, an open-source model suite from Nvidia that allows users to download, study, and customize models. He said the project is not designed to directly compete with OpenAI or Anthropic, but instead to strengthen the developer ecosystem closely tied to Nvidia’s hardware and software platform.
At the same time, Catanzaro said Nvidia “cannot stay out of the game.” As AI competition intensifies and is constrained by hardware supply, Nvidia needs to participate more deeply in shaping the ecosystem rather than relying primarily on partners to develop models and applications that drive chip demand.
Catanzaro framed Nemotron’s role as more than a research exercise, emphasizing that teams are competing for GPU resources. “This is not a science project. This is a matter of Nvidia's future,” he said.
For fiscal year 2026, Nvidia reported revenue of $215.9 billion, up 65% from the prior year and at a record high.
During its February earnings release, Nvidia did not forecast any data center compute revenue from China in its Q1 outlook due to the impact of U.S. export controls on advanced chips. Despite that, the company issued an upbeat quarterly forecast.
Nvidia projected revenue of around $78 billion for the quarter, allowing for a potential deviation of about 2%. This was above the $72 billion consensus on Wall Street.
Analysts said the stronger-than-expected revenue outlook could help ease concerns about a potential slowdown in AI infrastructure spending.
Premium gym chains are entering a “golden era” that is ending or already in decline, as rising operating costs collide with shifting consumer preferences toward more flexible, community-based ways to exercise. Long-term memberships are shrinking, margins are pressured by higher rents and facility expenses, and competition from smaller, more personalized…
