China's DeepSeek Launches Advanced AI Models, Challenging Global Tech Leaders

China's DeepSeek has unveiled preview versions of its latest AI chatbot models, DeepSeek-V4-Pro and DeepSeek-V4-Flash, a year after its initial model significantly disrupted the global tech scene. These new offerings aim to directly compete with major US AI developers such as OpenAI and Google. Following an open-source philosophy, DeepSeek provides unrestricted access to its source code for developers. The company claims the V4-Pro model leads all other open-source options in mathematics and coding, while closely trailing Google's Gemini 3.1-Pro in world knowledge. DeepSeek states the 'pro' version's performance is only slightly behind OpenAI's GPT-5.4 and Gemini 3.1-Pro, suggesting a development gap of three to six months from the most advanced frontier models. The 'flash' model offers similar reasoning capabilities to the 'pro' version but with enhanced speed and cost-efficiency. This release follows the impact of DeepSeek-R1, which last year drew comparisons to "AI's Sputnik moment" from venture capitalist Marc Andreessen. The emergence of DeepSeek continues to raise questions about data protection and potential governmental influence.