We've established a brand new company referred to as DeepSeek particularly for this purpose. This friend later based an organization price hundreds of billions of dollars, named DJI. However, LLMs closely depend on computational energy, algorithms, and data, requiring an preliminary investment of $50 million and tens of tens of millions of dollars per training session, making it tough for companies not value billions to sustain. There's still, now it's tons of of billions of dollars that China's putting into the semiconductor trade. Within the quantitative discipline, High-Flyer is a "prime fund" that has reached a scale of lots of of billions. Many startups have begun to adjust their methods or even consider withdrawing after main players entered the sphere, but this quantitative fund is forging ahead alone. 36Kr: Many imagine that for startups, entering the sector after main firms have established a consensus is not a very good timing. This means, by way of computational power alone, High-Flyer had secured its ticket to develop one thing like ChatGPT earlier than many main tech companies. Meta isn’t alone - other tech giants are also scrambling to understand how this Chinese startup has achieved such results. Meta is concerned DeepSeek outperforms its yet-to-be-released Llama 4, The knowledge reported.

"We are aware of and reviewing indications that DeepSeek may have inappropriately distilled our fashions, and can share data as we all know more," an OpenAI spokesperson said in a comment to CNN. In January 2025, a report highlighted that a DeepSeek database had been left uncovered, revealing over one million traces of sensitive information. When the shortage of excessive-performance GPU chips amongst home cloud suppliers grew to become the most direct factor limiting the start of China's generative AI, in keeping with "Caijing Eleven People (a Chinese media outlet)," there are no more than 5 companies in China with over 10,000 GPUs. China-targeted podcast and media platform ChinaTalk has already translated one interview with Liang after DeepSeek-V2 was released in 2024 (kudos to Jordan!) On this submit, I translated one other from May 2023, shortly after the DeepSeek’s founding. OpenAI, ByteDance, Alibaba, Zhipu AI, and Moonshot AI are among the many groups actively learning DeepSeek, Chinese media outlet TMTPost reported.

Nearly 20 months later, it’s fascinating to revisit Liang’s early views, which can hold the key behind how DeepSeek, despite restricted resources and compute access, has risen to stand shoulder-to-shoulder with the world’s leading AI corporations. In truth, this company, hardly ever considered by means of the lens of AI, has lengthy been a hidden AI big: in 2019, High-Flyer Quant established an AI company, with its self-developed deep studying coaching platform "Firefly One" totaling practically 200 million yuan in funding, equipped with 1,one hundred GPUs; two years later, "Firefly Two" increased its funding to 1 billion yuan, geared up with about 10,000 NVIDIA A100 graphics playing cards. General AI may be one of the next massive challenges, so for us, it is a matter of the way to do it, not why. Despite these challenges, High-Flyer remains optimistic. Within the swarm of LLM battles, High-Flyer stands out as the most unconventional participant. 36Kr: Are you planning to prepare a LLM yourselves, or give attention to a selected vertical industry-like finance-related LLMs? Since the discharge of its newest LLM DeepSeek-V3 and reasoning model DeepSeek-R1, the tech neighborhood has been abuzz with pleasure. Besides several leading tech giants, this list features a quantitative fund company named High-Flyer. Lots of the core members at High-Flyer come from an AI background.

DeepSeek CEO Liang Wenfeng, also the founding father of High-Flyer - a Chinese quantitative fund and DeepSeek’s main backer - lately met with Chinese Premier Li Qiang, the place he highlighted the challenges Chinese firms face due to U.S. After greater than a decade of entrepreneurship, this is the first public interview for this not often seen "tech geek" sort of founder. Therefore, past the inevitable matters of cash, talent, and computational energy concerned in LLMs, we also mentioned with High-Flyer founder Liang about what sort of organizational structure can foster innovation and how lengthy human madness can last. It bypasses security measures by embedding unsafe subjects among benign ones within a optimistic narrative. In May, High-Flyer named its new impartial organization devoted to LLMs "DeepSeek," emphasizing its deal with reaching truly human-degree AI. Liang Wenfeng: We won't prematurely design functions primarily based on models; we'll focus on the LLMs themselves. Liang Wenfeng: Our venture into LLMs is not immediately related to quantitative finance or finance basically. The more essential secret, perhaps, comes from High-Flyer's founder, Liang Wenfeng.