Chinese AI developer DeepSeek has reportedly raised approximately 100 billion yuan, or $15 billion, in one of the largest private technology financing rounds in China's history.
The financing was reported by Bloomberg on October 22, citing people familiar with the transaction. Reuters subsequently covered the report but stated that it could not independently verify the fundraising.
Battery manufacturer CATL and technology company Tencent were reportedly among the largest investors.
DeepSeek has not publicly confirmed the final financing amount, valuation or complete investor list in an official announcement.
According to Bloomberg, the company initially targeted approximately 50 billion yuan before increasing its fundraising goal to 80 billion yuan and eventually reaching a reported total of 100 billion yuan.
Founded in 2023 in Hangzhou by Liang Wenfeng, DeepSeek became one of the most closely watched AI laboratories after demonstrating that Chinese researchers could develop highly competitive large language models despite restrictions on access to advanced American semiconductor technology.
Liang previously co-founded High-Flyer, a quantitative investment firm that used machine learning and large-scale computing systems for financial trading.
DeepSeek developed its reputation through open-weight AI models, allowing developers and companies to run and adapt its technology without relying entirely on proprietary cloud APIs.
Its emergence intensified competition with American AI laboratories, including OpenAI, Anthropic and Google DeepMind.
The company has continued developing its model family while expanding investment in computing infrastructure and domestic technology partnerships.
The reported fundraising represents a substantial shift in the scale of DeepSeek's operations.
Training and operating frontier AI systems requires access to increasingly large computing clusters, specialized chips, high-bandwidth networking, electricity and engineering talent.
Even when research teams discover more efficient training methods, larger-scale experimentation and inference deployment can require considerable infrastructure investment.
The participation of industrial and technology companies, including Tencent and CATL, would also reflect the strategic importance of domestic AI capabilities to China's broader technology sector.
According to previous reporting, DeepSeek has been preparing for a potential domestic public listing. However, its final IPO timetable and valuation remain unconfirmed.
The fundraising also comes during a period of growing tension between the United States and China over advanced AI technology.
On October 19, the US Department of Homeland Security directed federal civilian agencies to remove DeepSeek-related technology from government IT systems, citing security and supply-chain concerns.
Those restrictions add to existing US measures affecting Chinese access to advanced semiconductor manufacturing equipment and AI computing infrastructure.
DeepSeek therefore faces an unusual combination of opportunities and constraints.
It has access to a large domestic market, growing investor interest and potential partnerships with major Chinese industrial groups. At the same time, hardware restrictions may increase the cost and complexity of building frontier AI infrastructure.
The company must also demonstrate that its open-weight model strategy can produce sustainable commercial revenue.
Distributing highly capable models can accelerate adoption, but monetization may depend on enterprise services, hosted inference, applications and other products built around those models.
If confirmed, the $15 billion financing would position DeepSeek among the most heavily financed AI developers globally.
For now, the amount should be treated as reported financing rather than a company-confirmed closing. The next significant disclosures will be the final capital structure, valuation, investor commitments and how the funds will be deployed.