Chinese artificial intelligence startup DeepSeek has officially launched its latest AI model, intensifying the global competition against American tech giants. The industry had been eagerly anticipating this announcement for weeks, viewing it as a definitive signal of China’s growing capabilities and ambitions in AI.
The company revealed the new DeepSeek-V4 series via its WeChat channel, offering the models as open-source resources. The series arrives in two distinct versions: the DeepSeek-V4-Pro, which offers enhanced reasoning, and the DeepSeek-V4-Flash, a more cost-effective option designed for efficiency and broader accessibility.
DeepSeek previously made waves in early 2025 with the release of its R1 agent. That launch caused significant volatility in United States technology stocks, as the startup demonstrated it could rival systems like ChatGPT, Gemini, or Claude at a fraction of the operational cost, surprising investors and analysts alike.
Read more about this topic: Tech industry watches closely as DeepSeek prepares new AI model
The launch arrives just one day after OpenAI announced its latest advancement, the GPT-5.5 model, which serves as the backbone for ChatGPT. With over a billion users, OpenAI’s announcement sets a high bar, ensuring the rivalry between Hangzhou’s rising star and Silicon Valley’s established leaders remains incredibly intense.
As the Hangzhou-based firm solidifies its position, the technology sector is closely monitoring how these new models will perform in real-world applications. The dual release of Pro and Flash versions suggests a strategy focused on balancing high performance with scalability, directly challenging the dominance of Western firms.
