Hangzhou DeepSeek Artificial Intelligence Basic Technology Research Co., Ltd.,[3][4][5] doing business as DeepSeek, is a Chinese artificial intelligence (AI) company that develops large language models (LLMs). Based in Hangzhou, Zhejiang, DeepSeek is owned and funded by High-Flyer, a Chinese hedge fund. DeepSeek was founded in July 2023 by Liang Wenfeng, the co-founder of High-Flyer, who also serves as the CEO for both of the companies.[7][8][9] The company launched an eponymous chatbot alongside its DeepSeek-R1 model in January 2025.
DeepSeek-R1 provided responses comparable to other contemporary large language models, such as OpenAI's GPT-4 and o1. Its training cost was reported to be significantly lower than other LLMs. The company claims that it trained its V3 model for US$6 million—far less than the US$100 million cost for OpenAI's GPT-4 in 2023[10]—and using approximately one-tenth the computing power consumed by Meta's comparable model, Llama 3.1.[10][11][12] DeepSeek's success against larger and more established rivals has been described as "upending AI".[13][14]
DeepSeek's models are described as "open-weight", meaning the exact parameters are openly shared, but the training data is not openly licensed.[15][16] Since the January 2025 debut of DeepSeek-R1, the company has made its new models available under free and open-source software licenses, primarily the MIT License.[17] The company reportedly recruits AI researchers from top Chinese universities[13] and also hires from outside traditional computer science fields to broaden its models' knowledge and capabilities.[11]
DeepSeek significantly reduced training expenses for their R1 model by incorporating techniques such as mixture of experts (MoE) layers.[18] The company also trained its models during ongoing trade restrictions on AI chip exports to China, using weaker AI chips intended for export and employing fewer units overall.[12][19] Observers say this breakthrough sent "shock waves" through the industry which were described as triggering a "Sputnik moment" for the US in the field of artificial intelligence, particularly due to its open-source, cost-effective, and high-performing AI models.[20][21][22] This threatened established AI hardware leaders such as Nvidia; Nvidia's share price dropped sharply, losing US$600 billion in market value, the largest single-company decline in U.S. stock market history.[23]
History
Founding and early years (2016–2023)
In February 2016, High-Flyer was co-founded by AI enthusiast Liang Wenfeng, who had been trading since the 2008 financial crisis while attending Zhejiang University.[25] The company began stock trading using a GPU-dependent deep learning model on 21 October 2016; before then, it had used CPU-based linear models. By the end of 2017, most of its trading was driven by AI.[26]
Liang established High-Flyer as a hedge fund focused on developing and using AI trading algorithms, and by 2021 the firm was using AI exclusively,[27] often using Nvidia chips.[28]
Company operation
DeepSeek is headquartered in Hangzhou, Zhejiang, and is owned and funded by High-Flyer. Its co-founder, Liang Wenfeng, serves as CEO. As of May 2024, Liang personally held an 84% stake in DeepSeek through two shell corporations.[50]
Strategy
DeepSeek has stated that it focuses on research and does not have immediate plans for commercialization.[51] This posture also means it can skirt certain provisions of China's AI regulations aimed at consumer-facing technologies.[11]
DeepSeek's hiring approach emphasizes skills over lengthy work experience, resulting in many hires fresh out of university.[32][11]
Training framework
High-Flyer/DeepSeek had operated at least two primary computing clusters: Fire-Flyer (萤火一号) and Fire-Flyer 2 (萤火二号). Fire-Flyer 1 was constructed in 2019 and was retired after 1.5 years of operation. Fire-Flyer 2 is still in operation as of 2025. Fire-Flyer 2 consists of co-designed software and hardware architecture. On the hardware side, Nvidia GPUs use 200 Gbps interconnects. The cluster is divided into two "zones", and the platform supports cross-zone tasks. The network topology was two fat trees, chosen for high bisection bandwidth. On the software side are:[55][26]
Distilled models were trained by SFT on 800K data synthesized from DeepSeek-R1, in a similar way as step 3. They were not trained with RL.[106]
There were reports that R2, the intended successor to R1, was originally planned for release in early May 2025.[107] However, on 28 May 2025, R1 was instead updated to version R1-0528.[108]
Significance
DeepSeek's success against larger and more established rivals was a surprise to both the industry and to markets,[13][113] and has been compared by investors and pundits to the "Sputnik moment".[13][114][115][22][21][20]
See also
- Reasoning model
- List of large language models
- Lists of open-source artificial intelligence software
- Zhejiang University
External links
References
- DeepSeek突传消息 Sina Corporation, 1 February 2025, retrieved 1 February 2025^
- Zijing Wu. DeepSeek focuses on research over revenue in contrast to Silicon Valley Financial Times, 14 March 2025, retrieved 14 March 2025^
- Hangzhou DeepSeek Artificial Intelligence Basic Technology Research Co., Ltd.