Advertisement
trendingNowenglish2850752https://zeenews.india.com/economy/who-is-liang-wenfeng-meet-founder-of-chinese-ai-chatbot-deepseek-rivaling-openais-chatgpt-2850752.html

Who Is Liang Wenfeng? Meet Founder Of Chinese AI Chatbot DeepSeek, Rivaling OpenAI's ChatGPT

After studying electronics and communication engineering at Zhejiang University, Liang Wenfeng graduated in 2010 with a master's in information and communication engineering.

Who Is Liang Wenfeng? Meet Founder Of Chinese AI Chatbot DeepSeek, Rivaling OpenAI's ChatGPT

New Delhi: DeepSeek, a Chinese AI business founded by Liang Wenfeng, has transformed the tech sector with effective models that compete with OpenAI and Anthropic.

Tech behemoths Nvidia, Meta, and Microsoft saw their shares plummet as a result of DeepSeek becoming the most downloaded and highly rated app on the Apple Store.

Who is Liang Wenfeng, founder of DeepSeek?
After studying electronics and communication engineering at Zhejiang University, Liang Wenfeng graduated in 2010 with a master's in information and communication engineering.

Liang managed more than 100 billion yuan by 2021 after co-founding a quantitative hedge fund in 2015. DeepSeek was founded a month later as the fund's focus moved to AI research in April 2023.

Liang remained primarily out of the spotlight, giving only two infrequent media interviews to the Chinese media source Waves in 2023 and last year. 

Media sources state that Liang was invited, along with other high-ranking officials and business leaders, to a closed-door symposium held by Chinese Premier Li Qiang on January 20.

What is DeepSeek? 
DeepSeek, a 2023-founded AI chatbot, looks and works like ChatGPT. In activities like mathematics and coding, it is said to be just as effective as OpenAI's o1 model.

The reasoning model R1 from DeepSeek reduces task costs, uses less memory, and simulates human thinking by producing responses incrementally.

DeepSeek's two models
The V3 model from DeepSeek, a potent large language model on par with OpenAI's GPT-4o and Anthropic's Claude 3.5, was released in December. The models can write essays, generate computer code, and respond to inquiries.

With the release of R1, a "reasoning" model, on January 20, DeepSeek improved its capacity to handle challenging issues including strategic planning and reading comprehension. Similar to OpenAI's o1, this model is a modified version of V3 that makes use of reinforcement learning.

Who can download DeepSeek? 
DeepSeek's models and techniques have been released under the free MIT License. Anyone can download it.

Great for the AI research community 
DeepSeek facilitates experimentation and development for the AI research community by giving access to enormous computing resources.

Access to AI cheaper 
With the ability to run more AI models on personal devices and do away with the requirement for cloud membership costs, consumers may be able to access AI at a lower cost.

Stay informed on all the latest news, real-time breaking news updates, and follow all the important headlines in india news and world News on Zee News.

NEWS ON ONE CLICK