DeepSeek set to raise at least $15 billion in Tencent-backed funding
Sign up now: Get ST's newsletters delivered to your inbox
DeepSeek’s funding sets the stage for one of the most highly anticipated Chinese market debuts in years.
PHOTO: REUTERS
- DeepSeek is set to raise at least 80 billion yuan (S$15 billion) in Tencent-backed funding, exceeding its initial target ahead of a planned IPO in early 2027.
- The company plans to build a massive AI data centre in Inner Mongolia using 160,000 Huawei AI chips, aiming to advance its large language model technology.
- DeepSeek's founder prioritises pushing AI technology boundaries over monetisation, reflecting China's ambition to compete globally in artificial intelligence.
AI generated
DeepSeek is close to securing at least 80 billion yuan (S$15 billion) in its latest round of funding, blowing past its own capital-raising target ahead of a landmark initial public offering in early 2027.
Battery pioneer Contemporary Amperex Technology (CATL) and WeChat-operator Tencent Holdings have committed among the largest amounts in the current financing, which closes soon, sources familiar with the matter said.
DeepSeek initially sought about 50 billion yuan, but interest outstripped expectations after the successful release of its latest AI model, the sources said, asking not to be named as the details are not public.
Based on term sheets signed, the final tally could approach 100 billion yuan, they added.
DeepSeek’s funding sets the stage for one of the most highly anticipated Chinese market debuts in years.
The company is riding high after the release of the V4 Flash large language model, which reset expectations for cost versus performance against leaders Anthropic and OpenAI.
That caps a strong run for founder Liang Wenfeng, whose start-up has forged a reputation for technical breakthroughs since a 2025 model that matched the capabilities of Silicon Valley offerings at a fraction of the cost.
As with other artificial intelligence players, DeepSeek requires huge amounts of capital to bankroll its ambitions in a potentially revolutionary technology.
The company plans to deploy at least 160,000 of Huawei Technologies’ top accelerators at a massive data centre it is building in Inner Mongolia, which may create one of the largest known clusters of Huawei AI chips.
It has publicly released software created with Huawei to program AI chips, a foray into a new arena in close partnership with the hardware leader.
Rival Moonshot, meanwhile, is also targeting an early 2027 IPO after closing funding at a US$50 billion (S$64 billion) valuation.
Liang pledged in at least one meeting with investors in 2026 that he will keep developing open AI models while pursuing a broader goal of achieving artificial general intelligence.
The hedge fund manager made it clear his main goal was to push the boundaries of the technology rather than monetisation, Bloomberg News has reported.
While discussions around the current deal are near an end, details may still change in final talks, the sources said.
DeepSeek had originally targeted a valuation of about 500 billion yuan in its financing.
A CATL representative declined to comment, while Tencent had no immediate comment.
DeepSeek did not respond to an e-mailed request for comment during a Chinese public holiday.
DeepSeek has drawn enormous interest from would-be investors because it is one of a clutch of companies that sit at the heart of China’s effort to compete globally on AI.
Once the round closes, the next step should be to restructure the company to prepare for an IPO, the sources said.
Tencent, which also landed the biggest investment in the start-up at 10 billion yuan in a previous round of financing, aims to integrate AI into its platforms.
CATL may be eyeing an increasingly lucrative business of supplying components to the industry, including for data centres.
The AI lab’s capital-raising had attracted its share of controversy. DeepSeek briefly suspended the deal after comments widely attributed to Liang about US-Chinese AI competition went viral on social media.
Chinese media outlet Yicai reported in July that the billionaire entrepreneur had talked about a reliance on Nvidia chips and China’s persistent lag in AI sophistication relative to the US.
Liang’s net worth more than doubled to around US$36 billion when DeepSeek closed its first round of external fund raising this summer.
The AI start-up grew as an offshoot from his hedge fund High-Flyer, which provided much of the initial support.
Liang contributed again in the previous 50 billion yuan round, which valued DeepSeek at roughly 350 billion yuan at the time. BLOOMBERG
