Liang Wenfeng Invests 20 Billion! DeepSeek's Record First Round Financing of 50 Billion, V4.1 Scheduled for June
DeepSeek aims to raise up to 50 billion yuan in its first funding round, with founder Liang Wenfeng personally contributing 20 billion. The company's valuation has surged to 350 billion yuan, and the V4.1 model is set for a June release, signaling a shift from an idealistic lab to a commercial AI company.
Article intelligence
Why it matters
This matters because deepSeek aims to raise up to 50 billion yuan in its first funding round, with founder Liang Wenfeng personally contributing 20 billion. The company's valuation has surged to 350 billion yuan, and the V4.1 model is set for a June release, signaling a shift from an idealistic lab to a commercial AI company.
Technical impact
May affect model selection, inference cost, product capability, and evaluation benchmarks.
DeepSeek, the Chinese AI startup known for its research-first approach, is making headlines with a record-breaking first funding round. The company aims to raise up to 50 billion yuan ($7 billion), with founder Liang Wenfeng personally contributing as much as 20 billion yuan, or 40% of the total. If successful, this would be the largest funding round ever for a Chinese large model company.
The funding round comes amid a rapid revaluation of DeepSeek. In just three weeks, the company's valuation quadrupled: from around $10 billion in early April 2026 to over $20 billion on April 22, then to about $45 billion on May 6, and potentially reaching $50 billion by early May. Investors include Tencent, Alibaba, and the National Integrated Circuit Industry Investment Fund.
DeepSeek had long been known for its reluctance to seek external funding or commercialization, focusing instead on research excellence. However, the AI landscape has shifted dramatically by 2026, forcing the company to confront three major realities: compute costs, talent retention, and productization.
First, compute demands are escalating. DeepSeek's V4 series, released in April, supports 1M context windows and is testing vision capabilities. Scaling up for enterprise use requires massive compute resources, not just for training but also for inference and reliable delivery. Second, the company has lost several star researchers to higher-paying competitors, highlighting the need for competitive compensation and equity incentives. Finally, DeepSeek is now actively marketing its models to enterprises, seeking to turn technology into revenue.
DeepSeek V4.1, expected in June, will further support this commercial pivot. It will offer more tools for enterprises, better integration with the industry-standard MCP protocol, and simultaneous processing of images and audio. The accelerated release cadence reflects the pressure from investors and the market.
In essence, this funding round marks a transformation of DeepSeek from a research lab into a capital-intensive AI company focused on compute, data centers, product teams, enterprise customers, and timely releases. While the deals are still under negotiation and DeepSeek has not commented, the direction is clear: the idealistic lab is learning to operate as a company.