A Chinese artificial intelligence company that caused global market turbulence last year has unveiled preview editions of its newest major technology upgrade on Friday, intensifying the competitive battle between China and the United States in the AI sector.
The V4 release from DeepSeek has been eagerly awaited by technology enthusiasts wanting to evaluate its performance against leading American AI systems including OpenAI’s ChatGPT, Anthropic’s Claude, and Google’s Gemini. Both Anthropic and OpenAI have previously claimed that DeepSeek improperly developed its technology using their innovations.
Technology industry experts had predicted the new model would debut over a month ago during the beginning of the Lunar New Year celebrations.
According to DeepSeek, the newly released V4 open-source models come in both “pro” and “flash” editions, featuring significant enhancements in knowledge processing, logical reasoning, and autonomous task execution capabilities – the ability to handle complex operations and workflows independently.
This V4 system follows the V3 model that DeepSeek introduced in the final months of 2024.
However, it was DeepSeek’s specialized reasoning AI system, known as R1, that surprised financial markets when it launched in January 2025. The company maintained it offered better cost efficiency than OpenAI’s comparable model and became representative of China’s progress in matching American technological innovation.
DeepSeek stated that its “V4 Pro Max” edition demonstrates “superior performance” on standard reasoning evaluations compared to OpenAI’s GPT-5.2 model and Google’s Gemini 3.0-Pro. The company acknowledged it performs “marginally” below GPT-5.4 and Gemini 3.1-Pro levels.
Regarding autonomous capabilities, the Chinese firm claimed its V4 “pro” edition could surpass Claude’s Sonnet 4.5 and nearly matches Claude’s Opus 4.5 model according to their internal testing.
The “flash” edition of V4 matches the “pro” version’s performance on basic automated tasks and demonstrates reasoning abilities that closely approach it, DeepSeek reported.
“Based on the benchmark results, it does appear DeepSeek V4 is going to be very competitive against its U.S. rivals,” said Lian Jye Su, chief analyst at the technology research and advisory group Omdia.
Marina Zhang, an associate professor at the University of Technology Sydney, described DeepSeek’s V4 launch as a “pivotal milestone for China’s AI industry”, particularly as worldwide competition grows more intense in the race for technological independence in essential technologies.
DeepSeek provides a free web-based and mobile chatbot service. Unlike leading models from Anthropic, Google and OpenAI, it characterizes its technology as “open source” by allowing developers access to modify and expand upon its fundamental technology.
Both V4 editions feature a 1 million token context window, which measures how much data an AI model can process and remember, and operate more efficiently, the company announced. This represents a substantial advancement from previous versions, as the V3 supported only a 128,000 token context window.
A Microsoft report from January revealed that DeepSeek usage has been expanding across numerous developing countries.
Nevertheless, some experts remain doubtful. Ivan Su, a senior equity analyst at Morningstar, described V4 as a “competent” successor but noted it doesn’t represent as significant an advancement as the R1 launch.
“Domestic competition has intensified significantly since R1’s release,” Su said. “Against U.S. models, DeepSeek’s own evaluation suggests its capabilities largely match on most fronts, but independent evaluations are needed before final conclusions can be drawn.”
In February, Anthropic accused DeepSeek and two additional China-based AI companies of conducting “industrial-scale campaigns” to “illicitly extract Claude’s capabilities to improve their own models.” The company said they accomplished this through a method called distillation that “involves training a less capable model on the outputs of a stronger one.” OpenAI made comparable accusations in correspondence to U.S. lawmakers.
This week, Michael Kratsios, chief science and technology adviser to U.S. President Donald Trump, also accused foreign technology companies “principally based in China” of distilling leading U.S. AI systems and “exploiting American expertise and innovation.”
China’s embassy in Washington responded to the accusations, calling them “unjustified suppression of Chinese companies by the U.S.”
