China’s AI landscape is rapidly evolving, and while companies like DeepSeek capture global attention, a group of less-publicized startups are quietly establishing their dominance. Known as the “Six Tigers” within the industry, these companies are spearheading innovation and shaping the future of artificial intelligence in the world’s second-largest economy.
Unveiling the “Six Tigers”
This elite group consists of Zhipu AI, Moonshot AI, MiniMax, Baichuan Intelligence, StepFun, and 01.AI. Each company contributes a unique set of strengths and specialized knowledge, but they all share a common characteristic: a highly skilled workforce with experience at leading tech corporations such as Google, Huawei, Microsoft, Baidu, and Tencent. These “tigers” are developing state-of-the-art AI models that directly compete with Western counterparts, expanding the possibilities in areas like natural language processing (NLP), computer vision, and multimodal AI. Their efforts are crucial in driving China’s technological advancement and asserting its presence in the global AI arena. The collective impact of these six companies is reshaping the landscape of AI development and deployment.
Zhipu AI: Pioneering Bilingual AI Innovation
Founded in 2019 by two professors from Tsinghua University, Zhipu AI distinguishes itself as one of China’s earliest pioneers in the field of bilingual AI. The company’s primary offerings include ChatGLM, a robust and versatile chatbot, and Ying, an AI-powered video creation tool designed to streamline and enhance video production processes. Zhipu AI’s early entry into the AI market has allowed it to establish a strong foundation and develop a deep understanding of the unique challenges and opportunities within the Chinese AI ecosystem.
In August of last year, Zhipu AI introduced its GLM-4-Plus model, which has garnered acclaim for its performance that rivals OpenAI’s GPT-4o. This achievement underscores Zhipu AI’s commitment to pushing the boundaries of AI technology and delivering models that can compete with the best in the world. The company also launched GLM-4-Voice, a conversational AI model capable of speaking both Chinese and English with natural-sounding intonations and regional accents. This capability represents a significant advancement in creating more human-like AI interactions, making it easier for users to communicate with AI systems in a natural and intuitive way. The ability to handle multiple languages and regional accents adds a layer of personalization and accessibility that is crucial for widespread adoption.
Despite its remarkable successes and technological advancements, Zhipu AI was recently added to the U.S. government’s trade restriction list. This development highlights the geopolitical complexities surrounding AI technology and the obstacles faced by Chinese companies operating in the global market. Despite these challenges, Zhipu AI secured over $140 million in funding earlier this year, with investments from Alibaba, Tencent, and several state-backed funds. This continued investment reflects a strong belief in the company’s long-term potential and its ability to navigate the challenges of the global AI market. The support from both private and public investors underscores the strategic importance of Zhipu AI in China’s broader AI strategy.
Moonshot AI: Revolutionizing Chatbot Interactions and Capabilities
Emerging from Tsinghua University in 2023, Moonshot AI was founded by Yang Zhilin, a researcher with a distinguished background from Carnegie Mellon University. The company’s standout product, the Kimi AI chatbot, has rapidly gained prominence, ranking among the top 5 most widely used chatbots in China. As of November 2023, Kimi AI boasted nearly 13 million monthly active users, according to Counterpoint Research. This rapid adoption rate highlights the chatbot’s appeal and its ability to meet the evolving needs of Chinese users.
One of Kimi AI’s key differentiators is its ability to handle queries of up to 2 million Chinese characters. This exceptional capacity allows users to engage in far more detailed and nuanced conversations compared to chatbots with more limited context windows. This is particularly useful for tasks that require extensive information processing, such as summarizing lengthy documents, translating complex texts, or generating creative content. The ability to process such large amounts of text enables Kimi AI to provide more comprehensive and accurate responses, making it a valuable tool for both personal and professional use. The extended context window allows for more complex and multi-faceted conversations, setting Kimi AI apart from its competitors.
With a valuation of $3.3 billion, Moonshot AI is backed by industry giants Alibaba and Tencent, further solidifying its position as a major player in the AI landscape. The backing of these tech giants not only provides Moonshot AI with significant financial resources but also gives it access to valuable expertise and infrastructure. This combination of innovation and strong financial support positions Moonshot AI to reshape the chatbot landscape and drive the future of conversational AI. The company’s focus on user experience and its ability to address real-world problems make it a compelling force in the AI market.
MiniMax: Crafting Immersive AI Characters and Experiences
Established in 2021 by AI researcher Yan Junjie, MiniMax focuses on creating AI-powered virtual characters that users can interact with. The company’s most notable product is Talkie, a chatbot platform that enables users to converse with a diverse range of AI personalities, spanning from celebrities to fictional characters. This innovative approach to AI allows users to explore different personas and engage in unique and entertaining conversations. MiniMax’s focus on creating immersive experiences sets it apart from traditional chatbot applications.
Originally launched as Glow in 2022, the app was later rebranded as Xingye in China and Talkie in international markets. This rebranding strategy reflects MiniMax’s ambition to expand its reach beyond China and establish a global presence. However, the Talkie app was removed from the U.S. App Store in December due to “technical reasons,” according to the South China Morning Post. Despite this setback, MiniMax continues to develop innovative AI solutions and adapt to the ever-changing landscape of the AI market. The company’s resilience and commitment to innovation are key factors in its continued success.
The company also offers Hailuo AI, a tool that generates videos from text prompts. This technology has potential applications in various fields, including marketing, education, and entertainment. The ability to generate videos from text opens up new possibilities for content creation and allows users to easily create engaging and informative videos. MiniMax achieved a valuation of $2.5 billion after a $600 million funding round led by Alibaba in March of last year. This significant investment underscores the company’s potential and its ability to disrupt the AI market. The backing of Alibaba provides MiniMax with the resources and expertise needed to further develop its AI technologies and expand its market reach.
Baichuan Intelligence: Championing Open-Source AI Development
Founded in March 2023, Baichuan Intelligence has assembled a team of experts with experience at Microsoft, Huawei, Baidu, and Tencent. This diverse team brings a wealth of knowledge and experience to the company, enabling it to develop cutting-edge AI technologies. Baichuan Intelligence is committed to promoting open-source AI development and has released two open-source language models, Baichuan-7B and Baichuan-13B, in 2023.
These models have been trained on multilingual data, making them versatile tools for a wide array of applications. They excel in areas such as general knowledge, mathematics, programming, translation, legal analysis, and healthcare. By making these models available to the public, Baichuan Intelligence aims to foster collaboration and accelerate innovation in the AI community. The open-source approach encourages developers to build upon and improve the models, leading to faster and more efficient AI development. The focus on multilingual capabilities ensures that the models can be used in a wide range of contexts and by users around the world.
Baichuan secured $687.6 million in funding in July, raising its valuation to over 20 billion yuan (approximately $2.8 billion). This funding round included participation from Alibaba, Tencent, and state-backed investment funds, demonstrating strong support for the company’s vision. The combination of private and public investment underscores the strategic importance of Baichuan Intelligence in China’s AI strategy. The company’s commitment to open-source development and its focus on practical applications make it a key player in the AI ecosystem.
StepFun: Pushing the Limits of Foundation Models
StepFun, an AI startup based in Shanghai, was founded in 2023 by Jiang Daxin, former Senior Vice President of Microsoft. Despite its relatively young age, the company has quickly made a name for itself by launching 11 foundation models, encompassing AI for image processing, audio processing, and multimodal applications. This rapid development and deployment of foundation models highlight StepFun’s agility and its ability to quickly capitalize on emerging opportunities in the AI market. The company’s focus on a wide range of AI applications demonstrates its commitment to addressing diverse needs and challenges.
Among these models, Step-2 stands out as a language model with a trillion parameters. This massive scale enables Step-2 to achieve impressive performance on various benchmarks. It currently ranks competitively with models from DeepSeek, Alibaba, and OpenAI on the LiveBench leaderboard, which provides real-time evaluations of large language model performance. The ability to compete with leading AI companies on performance benchmarks underscores StepFun’s technological prowess and its ability to develop state-of-the-art AI models. The focus on large language models with massive parameter counts reflects a trend in the AI industry towards increasingly complex and powerful models.
In December of last year, StepFun successfully raised hundreds of millions of dollars in a Series B funding round, backed by Fortera Capital, a state-owned private equity fund. This investment will fuel the company’s efforts to further develop its foundation models and expand its reach in the AI market. The backing of a state-owned private equity fund demonstrates the strategic importance of StepFun in China’s AI strategy. The company’s focus on foundation models and its ability to achieve competitive performance make it a key player in the AI ecosystem.
01.AI: The Vision of Kai-Fu Lee and Open Source AI
01.AI was founded in 2023 by Kai-Fu Lee, a veteran technology executive with experience at Apple, Microsoft, and Google. Lee’s extensive experience in the technology industry brings a wealth of knowledge and leadership to 01.AI. The company is a notable player in the open-source AI movement in China, with two primary models: Yi-Lightning and Yi-Large.
Both models have been released as open-source, and they have quickly risen to the top ranks of the world’s best-performing models in terms of language ability, reasoning, and contextual understanding. This commitment to open-source principles aligns with a broader trend in the AI community toward greater transparency and collaboration. The open-source approach allows researchers and developers to freely access, modify, and distribute the models, leading to faster and more efficient AI development. The focus on language ability, reasoning, and contextual understanding ensures that the models can be used in a wide range of applications, from natural language processing to knowledge representation.
A key feature of Yi-Lightning is its highly optimized training cost. According to Kai-Fu Lee, the model was trained using only 2,000 Nvidia H100 GPUs for one month, significantly less than xAI’s Grok 2, despite achieving comparable performance. This efficiency makes Yi-Lightning an attractive option for researchers and developers seeking to build powerful AI applications without incurring exorbitant costs. The ability to train a high-performing model with relatively low computational resources is a significant achievement and demonstrates 01.AI’s expertise in AI model optimization. This cost-effectiveness makes Yi-Lightning accessible to a wider range of users and organizations.
Yi-Large, on the other hand, is designed for natural human-like conversation, supporting both Chinese and English. Its ability to engage in fluent and contextually relevant dialogue makes it well-suited for applications such as customer service, virtual assistants, and interactive storytelling. The focus on natural human-like conversation makes Yi-Large a valuable tool for enhancing communication and interaction in various settings. The multilingual support ensures that the model can be used by a diverse range of users and organizations. The applications in customer service, virtual assistants, and interactive storytelling highlight the model’s versatility and its ability to address real-world needs.
A Force to Be Reckoned With in the Global AI Arena
These “Six Tigers” represent a significant force in China’s AI industry. Their innovative technologies, combined with strong financial backing and experienced teams, position them as key players in the global AI landscape. As they continue to develop and deploy cutting-edge AI solutions, they are poised to shape the future of AI in China and beyond. Their focus on open-source initiatives, bilingual capabilities, and multimodal applications reflects a diverse and dynamic approach to AI development. The collective impact of these companies is transforming the AI landscape and driving innovation in various sectors. Their ability to compete with Western counterparts and their commitment to addressing unique challenges in the Chinese market make them a force to be reckoned with in the global AI arena.