E-Paper
China's DeepSeek plans fresh fundraising at $74 billion valuation ahead of possible IPO
China's DeepSeek plans a new funding round at a $74 billion valuation, aims to raise up to 50 billion yuan, and is considering a Shanghai IPO.
Chinese AI startup DeepSeek is planning to raise more money from investors. The company could be valued at about 500 billion yuan ($74 billion) in this new funding round. It is raising the money before it may launch an Initial Public Offering (IPO), which is when a company starts selling its shares to the public on China's stock market.
DeepSeek is based in Hangzhou, China. The company became popular around the world in 2025 after launching low-cost AI models. These AI models got a lot of global attention. Just a few weeks ago, in June, Reuters reported that DeepSeek had already raised about $7.4 billion from investors. After that funding round, the company was valued at around 450 billion yuan.
DeepSeek is now planning another fundraising round, showing that investors are still eager to invest in the company. The company is looking to raise as much as 50 billion yuan in this new funding round. DeepSeek has also started early discussions about listing on Shanghai's STAR Market, which is China's Nasdaq-style stock exchange, according to three sources and two other people familiar with the plans.
The company has set an internal goal of filing IPO papers this year. The IPO and fundraising plans are still in the early stages. The amount to be raised, the valuation, and the timeline could still change. None of the people shared their names because the information has not been made public. Bloomberg News first reported that DeepSeek was preparing for a possible IPO filing. The Financial Times also reported that the company was considering a new fundraising round at a valuation of at least 480 billion yuan.
Why DeepSeek needs more money
DeepSeek surprised the global technology industry last year by launching AI models that competed with leading U.S. AI systems. Its AI models were seen as cheaper to train and operate than many competing AI models. Soon after its June fundraising, DeepSeek announced plans to double its workforce across different departments.
Also read: DeepSeek AI chip: China startup develops custom chip to reduce reliance on Nvidia and Huawei
The hiring will include teams working on data centres. The company also plans to expand work on AI agents, which are AI systems that can perform tasks with very little human instruction. Building data centres and developing advanced AI technology requires a lot of money. Earlier this month, Reuters reported that DeepSeek was working on developing its own AI inference chip. The company had quietly increased hiring of chip-design engineers for that project.
Change in DeepSeek's funding strategy
For a long time, DeepSeek avoided taking money from outside investors. The company was mainly funded by its founder Liang Wenfeng through his quantitative hedge fund High-Flyer, according to Reuters' earlier reporting. However, the cost of developing advanced AI has increased sharply. Because of these rising costs, DeepSeek has changed its strategy and is now raising money from outside investors.
Competition in China's AI market
DeepSeek is facing strong competition from major Chinese technology companies. Its competitors include ByteDance and Alibaba. It is also competing with well-funded AI startups such as Z.ai, Moonshot, and MiniMax.
Who invested in DeepSeek earlier
During the June fundraising round, founder Liang Wenfeng personally invested 20 billion yuan. Tencent Holdings invested 10 billion yuan. Battery maker CATL invested 5 billion yuan. Tencent and CATL became DeepSeek's largest outside shareholders after that funding round, according to Reuters.
Other investors include China's national AI fund, NetEase, JD.com, IDG Capital, Loyal Valley Capital, Monolith Management, and Shixiang Capital, according to sources and media reports. The investment by China's state-backed AI fund shows that Beijing considers DeepSeek an important company in its plan to build strong domestic AI businesses and reduce dependence on foreign technology.
