Zigpu and MiniMax, two members of the "Six Big Model Tigers," are now on the path to going public. With increasingly fierce competition, both companies are vying to become the "first large model listed company," a situation reminiscent of the previous "AI Four Little Dragons" listing boom. However, the market environment and business strategies they face this time are completely different.
First, the differences in their business models are evident. Zigpu mainly adopts the MaaS (Model as a Service) model, generating revenue through API calls, while MiniMax earns profits through subscription services for AI-native products. Zigpu tends to position itself as an API-driven revenue model, whereas MiniMax positions itself as a lightweight company centered around AI products. This difference in business positioning leads to clear distinctions in their style and market strategies.
In terms of market share, Zigpu and MiniMax also have different focuses. Zigpu concentrates on the domestic market, claiming to be first among independent general large model developers in China; MiniMax targets the global market, branding itself as the tenth-largest model company in the world. Nevertheless, the market share of both companies remains low, indicating that the "black hole effect" of the giants still exists.
Although both companies are achieving high growth, the logic behind their revenue growth rates differs. Zigpu's compound annual growth rate exceeds 130%, while MiniMax's growth is even more impressive, expected to reach 782.2%. Zigpu's revenue structure shows a shift towards cloud deployment, while MiniMax relies on the growth of its AI-native products.
However, the high growth rate also hides the pressure of "bloodied listing." Zigpu's losses exceed 6.2 billion yuan, and MiniMax's losses are about 9.3 billion yuan. From a cash flow perspective, MiniMax faces less short-term survival pressure and has stronger financial support.
In terms of talent strategy, Zigpu and MiniMax also have their own characteristics. Zigpu emphasizes the importance of a team of scientists, while MiniMax focuses on a younger organizational structure, showing different corporate cultures and talent management approaches. Although both companies have a high proportion of R&D staff, their per capita output capabilities differ significantly, with MiniMax's per capita output being three times that of Zigpu.
Finally, both companies have significant investments in computing power, reflecting the flow of funds in large model startups. In the future, Zigpu and MiniMax will continue to seek growth in the international market, while facing risks such as legal litigation and competition.
