AIbaseSource publication

Pouring in $15 billion! DeepSeek joins hands with CATL to head to Inner Mongolia, where a 1GW domestic-powered intelligent computing center…

According to multiple authoritative media reports, the artificial intelligence field has welcomed a high-profile massive investment. DeepSeek is recently close to completing a new funding round of up to $15 billion, and the core use of this capital has been clearly locked in — to invest in building a brand-new data center with a scale of 1GW in Ulanqab, Inner Mongolia.

According to multiple authoritative media reports, the artificial intelligence field has welcomed a high-profile massive investment. DeepSeek is recently close to completing a new funding round of up to $15 billion, and the core use of this capital has been clearly locked in — to invest in building a brand-new data center with a scale of 1GW in Ulanqab, Inner Mongolia.

In terms of the hardware compute foundation, the project demonstrates firm support for domestic chips. According to informed sources, the computing power of this large data center will be jointly provided by at least 160,000 Huawei Ascend 950DT chips, marking a key step forward for domestic large models in building ultra-large-scale intelligent computing clusters.

For a data center at the 1GW level, high power consumption and operating expenses are enormous challenges that no company can avoid. Ulanqab's uniquely abundant and inexpensive renewable green energy is the key to DeepSeek keeping operating costs down to a reasonable level.

It is worth noting that in this financing and infrastructure build-out, power battery and new energy giant CATL, as a major participating investor, added an additional $140 million in this round. CATL will not only provide capital support, but will also deeply empower the project, offering a complete end-to-end energy solution including large-scale energy storage systems, direct green power connections, and grid interfaces, ensuring a stable and clean power supply for the data center.

Industry analysts point out that as the project accelerates, the competitive landscape of the domestic artificial intelligence industry is undergoing profound changes. The industry's focus has gradually shifted from the early stage of "competing on whose model parameters are better" or "whose algorithms are smarter" to a new stage of competing on infrastructure, on the speed of data center construction, and on comprehensive energy guarantee capabilities.


Original source

AIbase

Content notes

Original publication and rights belong to the source.

Machine translation · Refer to the original