DeepSeek Closes Funding Round Worth Over $7 Billion, With Founder Investing the Most

DeepSeek Closes Funding Round Worth Over $7 Billion, With Founder Investing the Most

Ondřej Barták
Ondřej Barták
Entrepreneur and Programmer
17. 6. 2026
2 minutes reading · 8 views
DeepSeek Closes Funding Round Worth Over $7 Billion, With Founder Investing the Most

Chinese AI lab DeepSeek has just closed its first-ever external funding round. It raised roughly $7.4 billion, making it China's most valuable artificial intelligence startup. Investors valued the company at more than $50 billion. More interesting than the amount itself, however, is how the entire deal was structured.

The founder who refuses to relinquish control

Who put the most money into the company? Not a major fund or a technology giant. It was founder Liang Wenfeng himself, who invested around $3 billion in the round. That makes him by far the largest investor. Before the funding round, Liang held nearly 90 percent of DeepSeek. And he did everything he could to maintain his position.

Most external investors did not invest directly in DeepSeek. Instead, they channeled their money into a limited partnership controlled by Liang. This allowed the founder to retain control over the company's direction even as he brought in outside capital. He also added another condition. Investors must hold on to their stakes for five years and receive no voting rights. In other words, money is welcome, but influence over decision-making is not.

The only exception? China's National Artificial Intelligence Industry Investment Fund, a government-backed fund established last year. It invested roughly $150 million directly in DeepSeek and was the only investor to retain voting rights and remain exempt from the five-year lockup period. It had originally considered leading the entire round. In the end, it significantly scaled back its participation.

The company that took Silicon Valley by surprise last year

It is worth recalling where DeepSeek came from. It rose to global prominence early last year, when its V3 and R1 models won acclaim in Silicon Valley and challenged American assumptions about how far China had advanced in artificial intelligence.

DeepSeek unveiled a model that matched much more expensive Western systems at a fraction of the training cost. The result briefly rattled US technology stocks and changed thinking about how much computing power a cutting-edge model actually needs. Until then, the company had been virtually unknown outside a small circle of experts.

DeepSeek has since become something of a national champion. Beijing considers it important, the company's talented employees are subject to travel restrictions, and its technical achievements are viewed by the Chinese as a matter of national pride. A funding round centered on the founder and a handful of major domestic companies therefore also comes across as a strategic gesture.

Sources: theinformation.com and thenextweb.com

Advertisement

Content created with help from UpTier.

SEO and GEO on autopilot. UpTier’s multi-agent systems write and optimize content for search engines and AI answers.

Discover UpTier ↗

Category:AI
Did you enjoy this article?
Discover more interesting posts on our blog
Back to blog

Related posts

Google makes Tasks free for Gemini 4 Hydrogen usersGoogle makes Tasks free for Gemini 4 Hydrogen users
Tasks is becoming available for free to hundreds of millions of Gemini 4 Hydrogen users. Google says it can call businesses, book appointments and complete purchases on their behalf.
1 min read
18. 11. 2026
OpenAI introduces GPT-6.2 Astra and Sol for professional and everyday workOpenAI introduces GPT-6.2 Astra and Sol for professional and everyday work
OpenAI has introduced two models: Astra for complex professional tasks and Sol for everyday work. Both are available in ChatGPT Work, Codex and through the API.
3 min read
17. 11. 2026
Tencent releases Hy4 Flash with image support and a context window of up to 1 million tokensTencent releases Hy4 Flash with image support and a context window of up to 1 million tokens
The lighter Hy4 Flash model targets chat, documents and routine code edits. It accepts text and images, supports tool use and, according to Tencent, matched Claude Haiku 5.5 in tests of everyday tasks.
1 min read
16. 11. 2026
Přihlaste se k odběru našeho newsletteru
Zůstaňte informováni o nejnovějších příspěvcích, exkluzivních nabídkách, a aktualizacích.
CodedTrip

Operated by CodedTrip LLC, USA.

YouTube
TikTok