DeepSeek is preparing to raise about 50 billion yuan, or $7.4 billion, in its first outside funding round, according to people familiar with the matter. The deal could value the Chinese artificial intelligence startup at 350 billion yuan to 400 billion yuan, or about $52 billion to $59 billion, after the investment.
The planned round marks a shift for DeepSeek, which had long avoided outside capital. Founder Liang Wenfeng had funded the company through High-Flyer, his quantitative hedge fund. However, rising demand for advanced AI agents and computing power has pushed the company toward a larger capital base.
Tencent Holdings and CATL are expected to become the largest outside investors in the funding round. People familiar with the talks said Tencent is considering an investment of 10 billion yuan, while CATL is looking at 5 billion yuan. Liang is also expected to commit 20 billion yuan of his own money.
DeepSeek is also in final talks with China’s national artificial intelligence fund, NetEase, JD.com, IDG Capital, and Monolith Management. The planned investor group is expected to include fewer than 10 backers. However, the final size and investor list could still change before the round closes.
The fundraising places DeepSeek among China’s largest private technology deals. Still, it remains smaller than recent funding rounds by major US AI firms. OpenAI raised about $122 billion in March, while Anthropic raised $65 billion last month, according to the provided details.
DeepSeek, Liang, NetEase, JD.com, and the China Integrated Circuit Industry Investment Fund did not immediately respond to requests for comment. Tencent and CATL declined to comment, according to the information provided.
DeepSeek gained global attention early last year after its V3 and R1 models drew praise in Silicon Valley. The models challenged the view that top AI systems require the same huge budgets used by US rivals. Its rise also made the company one of China’s most closely watched AI startups.
However, the AI market has moved beyond low-cost, open-source chatbot models. Companies now focus more on AI agents, which can handle complex tasks with less human input. These systems often require more computing power, stronger infrastructure, and larger investment.
DeepSeek said in April that its next-generation V4 model, built around AI agents, redefined the state of the art for open-source models. Nevertheless, third-party evaluations suggest the model still trails some leading systems from US and Chinese competitors.
The planned funding round appears tied to that shift. More capital could support model development, computing resources, and the infrastructure needed to compete in agent-based AI. At the same time, DeepSeek faces limits that differ from those of US rivals.
US export controls remain a key factor in DeepSeek’s growth plan. Alfredo Montufar-Helu, a Beijing-based managing director at Ankura China Advisors, said Western restrictions limit the company’s access to advanced American chips.
“Western export bans mean DeepSeek cannot access frontier American silicon. Without the ability to buy that hardware, they have no reason to match the multi-billion-dollar computing budgets of their US rivals,” he said.
His comments show doubt over whether DeepSeek can follow the same spending path as OpenAI or Anthropic. Instead, the company may need to rely more on China’s domestic AI hardware, energy systems, and investor base.
The investor lineup also reflects China’s push to build a more self-sufficient AI industry. Tencent is working to grow its Hunyuan model, while Alibaba continues to promote Qwen. A closer tie with DeepSeek could help Tencent strengthen its position in China’s AI market.
CATL’s expected role also points to the rising link between AI and energy infrastructure. The battery maker has been exploring AI data centers, power equipment, and energy storage as AI workloads increase demand for reliable electricity.
Also Read: Startup News: China Blocks Meta’s $2B Acquisition of AI Startup Manus