Summary of Key Points
The AI large model developed by DeepSeek recently reached 7.22 trillion token requests in a single week, surpassing established AI companies like OpenAI and becoming the most utilized AI model globally. This indicates that it has temporarily taken the lead in terms of user adoption, reflecting its growing popularity and competitiveness in the market.
Detailed Analysis
1. Understanding Terms: What are Requests and Tokens? What Does 7.22 Trillion Mean?
- Requests: Simply put, these represent the number of times users have used the AI model or the amount of data processed. For example, every time you use the model to write an article, chat, or code, it counts as a request. The more content is processed, the higher the number of requests.
- Tokens: These are the smallest units used by AI models to process text—each Chinese character, English word (or part of a word), etc., counts as one token. For instance, “你好” (Hello) consists of 2 tokens, while “hello” contains 1 token.
- The Significance of 7.22 Trillion: This is an enormous number, suggesting that millions of users around the world are using DeepSeek to process large amounts of content daily—perhaps for writing reports, translating documents, or generating code. This figure clearly demonstrates that DeepSeek’s user base has surpassed that of OpenAI (the model behind ChatGPT).
2. How Did DeepSeek Overcome OpenAI and Other Competitors?
OpenAI was the pioneer in AI large models, so why was it surpassed by a newcomer? There could be several reasons:
- Cost Advantage: DeepSeek’s services may be more affordable. For example, if processing 1000 tokens costs less than with OpenAI, both businesses and individual users are more likely to choose DeepSeek.
- Better Adaptation to Chinese Context: As a model developed by a Chinese company, DeepSeek may handle Chinese language tasks (such as writing text or analyzing documents) more effectively, attracting many domestic users.
- Targeted Features: DeepSeek might have been optimized for specific use cases like code generation and mathematical calculations, appealing to programmers and students. It also supports multiple languages and industries (e.g., finance, healthcare).
- Market Promotion and Ecosystem: DeepSeek may have been more actively promoted in China and certain overseas markets (e.g., Southeast Asia), collaborating with businesses to provide customized AI services, resulting in a steady stream of requests.
3. What Does This “Global Leadership” Mean for the AI Industry?
- Breaking Monopolies: Previously, OpenAI was almost the dominant force in the AI industry. Now that DeepSeek has emerged, it shows that Chinese AI companies are capable of competing with established players on a global scale and are no longer just followers.
- Intensified Competition: This will encourage companies like OpenAI and Google to improve their products by lowering prices, enhancing features, or launching models suitable for different regions, ultimately benefiting users.
- Rise of Chinese AI: This marks the advancement of China’s AI technology, indicating that we are at the forefront in large model research and application. More Chinese AI companies may emerge on the global market in the future.
4. What Benefits Do Ordinary Users Enjoy?
- Cheaper AI Services: With increased competition, prices for AI tools are likely to decrease, making it more affordable for users.
- Better Chinese Language Support: If DeepSeek continues to improve its handling of the Chinese language, using it for writing or chatting will be more seamless, avoiding misunderstandings that some foreign models may have.
- More Choices: With multiple options available (e.g., DeepSeek for coding, OpenAI for English conversations), users can choose the best fit for their needs based on cost and functionality.
- Faster Technological Progress: Competition will drive companies to innovate faster, leading to higher-quality AI-generated content, faster response times, and new features (such as AI-generated images or videos).
5. Any Concerns?
Although leading in requests is impressive, quality is also important. It’s worth considering whether users are using DeepSeek for simple tasks (e.g., chatting) or complex ones (e.g., scientific research). For many simple tasks, high request volumes may not equate to significant value per user. Additionally, this data is based on a single week; it’s uncertain whether DeepSeek can maintain its lead in the long term, as OpenAI could soon release new models or lower prices.
Regardless, this development indicates that the AI industry is becoming increasingly competitive, which is great news for all of us. DeepSeek’s success is not only a testament to its own capabilities but also a reflection of the rise of China’s AI ecosystem, making the global AI market more dynamic and offering us better and more affordable AI tools in the future.