Tencent, Alibaba in talks to invest in DeepSeek at $20 billion-plus valuation

Tencent and Alibaba are in talks to invest in DeepSeek as the AI startup pursues its first outside funding round at a valuation north of $20 billion, according to a report by The Information.

The company is seeking to raise at least $300 million, marking a sharp shift for a startup that has so far relied entirely on backing from its parent, High-Flyer Capital Management. For most of its short life, DeepSeek has operated without external pressure, focusing on research rather than revenue growth.

“Chinese tech giants Tencent Holdings and Alibaba Group are in talks to invest in DeepSeek, the AI upstart that recently started fundraising for the first time,” The Information reported, citing four people with knowledge of the discussions

“DeepSeek, owned by Chinese hedge fund High-Flyer Capital Management, is now seeking to raise funding at a valuation of more than $20 billion, after its initial discussions with prospective investors generated immense interest,” The Information added, citing one of the people.

The Information first broke the news Friday that DeepSeek has begun talks to raise outside capital for the first time, aiming to secure at least $300 million at a valuation of $10 billion or more.

DeepSeek’s Next Move: $20B+ Valuation Talks with Tencent and Alibaba Signal AI Power Shift

Demand for DeepSeek’s models and API services has surged, pushing its infrastructure to the edge. More usage means more compute, more servers, and rising costs to keep systems running. Training and serving advanced reasoning models and emerging agent-style AI requires serious capital, even for a team known for efficiency.

“DeepSeek is in talks to raise outside capital for the first time, seeking to beef up its financial war chest so it can better compete in the costly battle to develop leading AI models,” The Information reported, citing four people familiar with the matter.

The move signals a new phase. DeepSeek built its reputation as a research-first lab, not a startup chasing rapid commercialization. Now it is facing the same pressures as every major AI player: scale.

The company’s breakout came with the release of its R1 model in early 2025. The performance drew comparisons to leading systems from U.S. labs. What stood out was the cost. R1 was reportedly trained for about $5.6-$6 million using older Nvidia H800 chips, far below what competitors typically spend.

That number shook the market.

Investors began questioning whether massive budgets were the only way forward. Chip stocks slipped. Momentum across AI names cooled. At one point, estimates suggested close to $1 trillion in value was erased across AI-related companies.

DeepSeek leaned into a different approach. Its models are open-source, and its inference costs are lower in certain use cases. The team focused on techniques such as KV cache compression and selective activation, demonstrating how constraints can drive smarter design.

Growth has exposed limits.

Usage spikes have led to outages. Infrastructure costs are rising. High-Flyer’s strong performance has supported DeepSeek so far, with reports pointing to returns of about 56.6% in 2025. That backing gave the company room to operate independently. The gap between demand and capacity is widening.

Founder and CEO Liang Wenfeng has resisted outside investors, concerned that external pressure could reshape the company’s priorities. That stance helped define DeepSeek’s identity.

This funding effort signals a recalibration, not a full shift.

Valuation discussions reflect uncertainty around the company. Estimates have ranged from as low as $1 billion to well above $20 billion. The current target sits in the middle, balancing its influence with its relatively lean structure.

Who invests matters. Chinese investors are the most likely participants. U.S. firms face regulatory pressure and national security concerns that could limit involvement. Export controls on advanced chips continue to shape how DeepSeek builds and scales.

The stakes go beyond one company.

The AI race is no longer just about building larger models. It is about efficiency, infrastructure control, and the ability to sustain rising costs.

DeepSeek forced that conversation. Now it has to prove it can compete at the next level.

If the round closes, the startup would join the ranks of China’s most valuable AI companies. The capital would help expand infrastructure, stabilize services, and advance commercialization.

For a company that showed AI breakthroughs do not always require massive budgets, the next test is clear: handling scale without losing what made it different.