DeepSeek's latest funding round is wrapping up. Reuters reported on October 22, 2026, citing two people with knowledge of the matter, that the round raised about 100 billion yuan, roughly $14.98 billion, and is set to complete in the coming days; Bloomberg first reported the roughly $15 billion total the same day. Battery maker CATL and Tencent were among the biggest investors, with CATL reportedly contributing about $1.5 billion.
From a 50 billion yuan target to about 100 billion
The round began with a target of about 50 billion yuan and was raised at least twice during talks, first to about 80 billion yuan and finally to about 100 billion yuan — double the original plan, at a pre-money valuation of around $74.8 billion. Together with a first external round of more than $7.5 billion closed in June, DeepSeek has raised roughly $22.5 billion in about six months. Reuters also cautioned that the financial details are not yet ironclad and could still change, and none of DeepSeek, CATL or Tencent commented — so these figures are a reported closing size, not a company announcement.
Where the money goes, and where the IPO stands
Earlier reporting sketched the spending plan: building its own data centers, expanding inference compute, and funding larger model training and open-weight releases. For a lab that built its name on low-priced APIs and open weights, a doubled raise changes the playbook — low prices need real compute underneath, and spending starts to resemble that of its US rivals. The listing track is moving too: Reuters reported in September that DeepSeek had hired CITIC Securities to prepare for a potential STAR Market IPO. Once this round closes, the market's question shifts from how much it can raise to whether models and revenue keep pace with the spending.
