DeepSeek’s latest flagship model, V4 Pro, has not managed to surpass its domestic and US competitors, struggling to replicate the groundbreaking success of its earlier R1 release. Despite being the most advanced system from the Chinese AI firm, V4 Pro secured the second position among top open-source models globally, trailing behind Beijing-based Moonshot AI’s Kimi K2.6.
Although V4 Pro shows clear improvements over its predecessor, V3.2, it still falls short compared to leading closed-source models from the US, such as OpenAI’s GPT-5.5, Anthropic’s Claude Opus, and Google’s Gemini 3.1 Pro. These models achieved scores of 60, 57, and 57 respectively on the Artificial Analysis Intelligence Index, while V4 Pro scored 52 compared to Kimi K2.6’s 54.
These findings underscore the challenges DeepSeek faces as China strives to close the AI gap with the US, amid fierce competition both domestically and internationally, as well as limitations in computing power. Nonetheless, analysts acknowledge that V4 Pro represents significant technical advancement.
