DeepSeek, the AI startup that rattled global markets from a Hangzhou server room, is raising outside funding for the very first time. The valuation? Above $20 billion. And both Tencent and Alibaba want in.
A week ago the valuation was $10 billion. Then the investors showed up. Within days it reportedly doubled.
Tencent, the Chinese multinational technology conglomerate, has reportedly pushed for a significant stake (as much as 20% according to some reports), with DeepSeek pushing back on ceding that level of control. That is not a company scrambling for cash. That is a company that knows its worth.
It’s this level of confidence that has a $20 billion price tag on it now. And China’s big tech is racing to have a slice.
From $10 Billion to $20 Billion in a Matter of Days
Investor demand arrived so fast and so strongly that the target valuation climbed past $20 billion before the round had even formally opened.
That kind of upward revision does not happen without serious momentum behind it. It tells you the people studying this deal believe they are looking at something genuinely valuable, not just a story that got loud.
Tencent and Alibaba: The Names That Want In
Tencent has reportedly proposed taking as much as a 20% stake in the company as part of the funding round. DeepSeek is said to be pushing back on ceding that level of control.
That level of detail matters. A company that holds firm against one of China’s most powerful and well capitalised technology businesses is a company that knows exactly where it stands. It is not scrambling for capital. It is choosing its partners.
Alibaba is also in discussions, though the precise terms it is proposing have not been confirmed. Both companies are in a race to build out their AI capabilities and infrastructure, and a stake in DeepSeek would give either of them exposure to one of the most credible AI research operations running anywhere in the world right now.
Why the Open Source Bet Is Paying Off
Built by Chinese hedge fund High-Flyer Capital Management, DeepSeek is an AI research company that has managed to match the output of the world’s best funded labs while spending a fraction of what they do. It sits in an unusual position because of it.
Its models have been released as open source, meaning any developer, researcher or company can use or build on them without restriction. That approach has sparked debate about how you actually value a business like this.
But the open source bet has not weakened DeepSeek. It has made it more visible, more embedded in global AI development, and more trusted than a closed model could ever become in the same time. The valuation question is a genuinely hard one to answer. The strategic value is not.
The Numbers Around It Tell a Clear Story
To put the $20 billion figure in context, look at the nearest comparable market participants. MiniMax and Zhipu, two of DeepSeek’s Chinese competitors, both listed in Hong Kong in January at valuations under $10 billion. MiniMax has since climbed above $30 billion. Zhipu has passed $50 billion. Moonshot AI, the company behind the Kimi series of large language models, is currently seeking $18 billion in its own funding round.
Against that peer group, a $20 billion entry point for DeepSeek starts to look conservative.
Tencent’s proposed benchmark for the valuation is MiniMax at around $40 billion. If that framing holds, there is a reasonable argument the final number ends up considerably higher than where it sits today.
What This Round Actually Means
This is more than a funding story. It is a validation story.
DeepSeek never needed outside money before. It was funded internally by High-Flyer and built its reputation without a single outside investor. What has changed is the scale of ambition. Running frontier AI models, expanding research teams and competing at the highest level globally is expensive in a way that internal funding eventually cannot cover alone.
Ultimately, the Chinese company built competitive AI models at a fraction of what Western counterparts spend. It proved that world class research does not require world class budgets. The first outside fundraise turns that proof of concept into a formal structure. Investors, partners and users now have cleaner visibility into what the company is and where it is heading.
And it is happening inside a Chinese AI race that is accelerating faster than most people expected. The capital is moving. The ambition is scaling. DeepSeek just put a number on what that looks like.
See Also:
Why TSMC’s Record Profits Point to a Bright AI Future
