DeepSeek
Leadership History
1 leader · 2023 to present
Liang Wenfeng is the CEO of DeepSeek, in the role since 2023.
| Name | Role | From | To | Years in role |
|---|---|---|---|---|
| Liang Wenfeng | Founder & Chief Executive Officer | 2023 | Present | 3 |
DeepSeek is a small, research-led Chinese AI company headquartered in Hangzhou. Instead of competing on the size of its GPU fleet, it competes on efficiency: sparse Mixture-of-Experts models, Multi-head Latent Attention to shrink inference memory, FP8 training and reinforcement-learning recipes for reasoning. With roughly 160 employees in 2025, it produced models that sat near the top of global benchmarks and the most downloaded app on the US App Store in January 2025. In 2026 the company moved from a hedge-fund-backed lab toward a commercial company, with a ~$7.4 billion Series A, an IPO plan and the V4 model family. The company is headquartered in Hangzhou, China. The incredibly secretive Chinese artificial intelligence lab was founded in Beijing in 2023, aggressively attempting to catch up to OpenAI and Google by releasing massive open-source language models fueled by immense compute resources.
DeepSeek makes money mainly by selling access to its models through a pay-per-token API (OpenAI-compatible endpoints such as deepseek-v4-pro and deepseek-v4-flash), priced well below US rivals. Its consumer chatbot on web and mobile is free. Model weights are published under the MIT license, so anyone can self-host them; this spreads adoption and pulls developers toward DeepSeek's hosted API, while cloud providers such as Microsoft Azure also host the models. DeepSeek does not disclose revenue. Until 2026 it was funded almost entirely by High-Flyer and Liang Wenfeng; the June 2026 Series A brought in Tencent and CATL as its largest outside shareholders.
Founder & Chief Executive Officer
2023 – Present
3 years (current)
Liang decided to aggressively open-source DeepSeek's frontier models (like DeepSeek-V2) and significantly undercut OpenAI's API pricing to rapidly capture developer market share.
As CEO, he utilized the massive, pre-existing compute infrastructure from his high-frequency trading firm to rapidly iterate on LLM architectures, bypassing the traditional venture capital funding cycle.
DeepSeek's main constraints are US export controls that block access to Nvidia's most advanced GPUs, government bans or restrictions on its app and data practices in several countries (including Italy, South Korea, Australia and US federal agencies in 2025), and accusations from US labs, notably Anthropic in February 2026, that it trained on outputs from their models via fraudulent accounts. Serving peak demand with limited compute has also caused outages.
Liang Wenfeng is the Founder of DeepSeek and Founder/CEO of High-Flyer Capital Management. A graduate of Zhejiang University with degrees in Electronic Information Engineering and Computer Science, Liang is a pioneer in algorithmic quantitative trading in China who built High-Flyer into a premier quantitative fund before launching DeepSeek in 2023.
In 2019 to 2021, Liang Wenfeng built High-Flyer's dedicated AI supercomputer 'Do pond' (housing 10,000+ GPUs) to train financial predictive models. Recognizing that general artificial intelligence was the ultimate frontier of computing, Liang launched DeepSeek in 2023 with a dedicated mission toward open AGI.
Liang Wenfeng champions a culture of extreme algorithmic purism, flat research hierarchies, first-principles mathematical thinking, and long-term intellectual curiosity, eschewing corporate commercial pressure to focus 100% on breakthrough machine learning research.
DeepSeek's research team consists of top young Chinese mathematicians, competitive Olympiad medalists, and systems engineers from top universities (Tsinghua, Peking, Zhejiang University) who lead core innovations in reinforcement learning, MLA attention, and distributed systems.
Because DeepSeek is funded directly by Liang Wenfeng and High-Flyer without external venture capital, the company is free from commercialization pressure, allowing it to release models completely open-source and price APIs near break-even costs.
Liang Wenfeng envisions a future where frontier AGI is an open, universal public good accessible to all humanity, driven by transparent, reproducible scientific research rather than closed corporate monopolies.
DeepSeek operates like an academic institute: researchers have unlimited compute access, work in small autonomous project pods, engage in daily paper discussions, and focus on fundamental algorithmic design rather than administrative tasks.
Following the release of DeepSeek-V3 and DeepSeek-R1, Liang Wenfeng and DeepSeek gained international acclaim, recognized by global AI luminaries for proving that algorithmic ingenuity can outcompete large brute-force compute budgets.
Liang Wenfeng is the primary driver behind DeepSeek. He is the founder of High-Flyer Quant, notoriously media-shy, and recognized that the same deep learning architectures used to predict massive stock market movements could be adapted to build world-class language models.
Compare executive leadership tenures and strategic transitions
Explore DeepSeek
Core profile pages, founders and related research hubs for this company.