# Сопутствующие статьи по теме Capacity

Новостной центр HTX предлагает последние статьи и углубленный анализ по "Capacity", охватывающие рыночные тренды, новости проектов, развитие технологий и политику регулирования в криптоиндустрии.

DeepSeek Paralyzed for 12 Hours: Is the Computing Power of Domestic Large Models Failing to Keep Up with Ambitions?

On the evening of March 29, DeepSeek, a leading Chinese large language model, experienced a 12-hour service outage affecting both its web and app platforms. The disruption, marked by repeated server failures and instability, raised significant concerns about the platform's reliability. Initial explanations pointed to overwhelming user traffic, but data did not support a sudden surge in users. Instead, the incident highlighted deeper structural challenges, particularly in computational supply struggling to keep pace with growing AI demands. As models become more complex—supporting longer context, advanced reasoning, and multimodal tasks—the computational load increases substantially. The outage also underscored emerging usage patterns like “lobster farming,” where automated, high-frequency API calls amplify server load. Meanwhile, anticipation for DeepSeek’s upcoming V4 model—featuring major upgrades like million-token context and stronger multimodal capabilities—suggests even greater future computational pressure. This incident signals a shift in the AI industry: competition is no longer just about model capability but also infrastructure stability, cost efficiency, and engineering scalability. The outage serves as an early warning of the systemic challenges facing AI platforms as adoption expands.

marsbit04/03 12:23

DeepSeek Paralyzed for 12 Hours: Is the Computing Power of Domestic Large Models Failing to Keep Up with Ambitions?

marsbit04/03 12:23

$25 Billion: Tesla Buys the Lowest-Tier Entry Ticket to the Chip Arms Race

Elon Musk has announced Tesla's plan to invest approximately $25 billion to build a semiconductor superfab named "Terafab," targeting 2nm process technology with a production capacity of 100,000 wafers per month. The move aims to address Tesla's soaring demand for AI chips, driven by its autonomous driving systems, Optimus robots, and upcoming Robotaxi fleet, which existing foundries like TSMC and Samsung cannot fully support. However, the $25 billion budget is considered insufficient by industry standards. For comparison, TSMC’s Arizona fab costs $165 billion, Samsung’s Taylor fab $44 billion, and Intel’s Ohio project $28 billion. A standard 2nm fab with 50,000 wafers/month typically requires around $28 billion, meaning Tesla’s goal is highly ambitious. Tesla’s chip development has been rapid: from HW3 (14nm, 144 TOPS) to AI5 (3/2nm, 2000+ TOPS), with performance multiplying every generation. Its growing reliance on external foundries led to a $16.5 billion long-term deal with Samsung for AI6 production. Terafab represents a natural shift toward self-sufficiency. The project faces significant challenges, including a 3–5 year construction period and additional time for production ramp-up. If Tesla follows industry timelines, Terafab may not be operational until 2029–2030, coinciding with expected mass production of Optimus and Robotaxi. Musk has also hinted at potential collaboration with Intel, which has advanced 18A process capacity. The $25 billion investment buys Tesla a entry ticket into semiconductor manufacturing—but whether it becomes a milestone in vertical integration or an overambitious project remains to be seen.

marsbit03/16 11:06

$25 Billion: Tesla Buys the Lowest-Tier Entry Ticket to the Chip Arms Race

marsbit03/16 11:06

活动图片

Популярные категорииright-arrow

Популярные тегиright-arrow

Bitcoin

Indepth Research