DeepSeek Pauses 480 Billion Yuan Funding Round After Founder's Remarks Leak
DeepSeek has paused its second funding round, which targeted a valuation of about 480 billion yuan (roughly $66 billion). The reason is a leak of founder Liang Wenfeng's comments from closed investor meetings, which spread across Chinese social media and went viral. According to Bloomberg, Liang himself verbally asked prospective backers to hold off on the round.
AI-processed from TNW; edited by Hamidun News
DeepSeek has paused its second fundraising round, Bloomberg reported on July 24, 2026. Founder Liang Wenfeng verbally asked prospective investors to hold off on committing funds after his comments from closed meetings leaked and went viral on Chinese social media. The round targeted a pre-money valuation of about 480 billion yuan.
Why DeepSeek put the round on pause
DeepSeek halted fundraising because of a leak: Liang Wenfeng's remarks, made in private meetings with prospective backers, reached Chinese social media and spread virally. According to Bloomberg, it was only after this that the founder personally and verbally asked investors to hold off on participating. The company has not formally announced that the deal is closed or canceled.
- Bloomberg published its report on July 24, 2026
- The round's target pre-money valuation is about 480 billion yuan
- The reason for the pause is a leak of founder Liang Wenfeng's comments
- The comments were made at closed meetings with investors
- The initiative to pause came from Liang himself
Liang
Wenfeng verbally asked prospective investors to hold off on the round.
— according to Bloomberg, July 24, 2026
How much DeepSeek wanted to raise
The round targeted a pre-money valuation of about 480 billion yuan, or roughly $66 billion at the current exchange rate (about 7.2–7.3 yuan per dollar). Pre-money valuation means the company's value before new capital comes in, and a figure like this would have put DeepSeek alongside the world's largest private AI companies. Bloomberg does not disclose the exact amount to be raised or the list of investors. The only publicly known details are the target valuation and the fact that this was the lab's second round.
What DeepSeek is known for
DeepSeek is a Chinese AI lab that grew out of the quantitative hedge fund High-Flyer; its founder is the same Liang Wenfeng. The company gained global fame in January 2025, when the release of its open model DeepSeek-R1 showed performance comparable to OpenAI's products and triggered a sharp drop in US tech stocks, including Nvidia. Since then, the company's valuation and investor interest in it have grown steadily.
That is why the pause in the second round is a notable event: this is one of the most valuable private AI companies, and the reason for the halt lies not in the business numbers but in a reputational incident involving the founder's words.
What this means
For the market, this is a signal of how fragile mega-rounds for AI companies have become: a single leak of private comments can put a deal valued at hundreds of billions of yuan on hold. So far, DeepSeek has announced neither a new timeline nor a cancellation. The fundraising status remains open, and the future of the 480-billion-yuan deal depends on how the company handles the reputational blow.
FAQ
Why did DeepSeek pause the round?
Because of a leak: comments by founder Liang Wenfeng from closed investor meetings reached Chinese social media and went viral. After that, according to Bloomberg, Liang himself asked backers to hold off on the round.
What was the valuation of DeepSeek's round?
The round targeted a pre-money valuation of about 480 billion yuan, roughly $66 billion at the current exchange rate. It was the company's second funding round.
Who is Liang Wenfeng?
Liang Wenfeng is the founder of DeepSeek and of High-Flyer, the quant fund affiliated with it. His public and private statements directly shape how investors perceive the company, as the current incident confirms.
Access Frontier LLMs & Multimodal AI in One Workspace
Discover NomadNet AI (nomadnet.ai) — unified access to frontier language models, research synthesis, and multimodal workflows.
The AI world, distilled — once a week
Seven stories that actually mattered, hand-picked. No noise, no reposts, no press releases.
Done! Check your inbox for a confirmation.