DeepSeek, the Chinese AI startup that surprised the world with its low-cost models last year, has offered a preview of its much-awaited V4 series, released this Friday.
The V4 Pro version is ahead of other open-source models in world knowledge, trailing only Google’s Gemini 3.1 Pro.
The company also launched V4 Flash, a more affordable version designed to incorporate real-world user feedback.
At the moment, no specific timeline has been released for the model, which is expected to be finalized in the near future. According to a report released by The Information, the main objective of DeepSeek is to raise funds at a valuation of more than $20 billion.
Despite the controversy, DeepSeek is reportedly seeking a valuation of more than $20 billion, with tech giants Alibaba and Tencent expected to acquire stakes.
Meanwhile, the White House has accused China of stealing intellectual property from US AI labs on an industrial scale – a move that threatens to disrupt an upcoming summit between US and Chinese leaders.
DeepSeek has faced repeated allegations from the US and its allies regarding inappropriate and illegal conduct.
The Chinese Embassy in Washington dismissed the claims as “baseless allegations”, saying Beijing is committed to protecting intellectual property rights.
Huawei has worked closely with DeepSeek to ensure that the new V4 models are fully compatible with its high-performance Ascend AI system.
Huawei’s entire Ascend Supernode product line now supports the DeepSeek v4 series, indicating deeper integration of Chinese hardware and software.
Following US sanctions on advanced AI chips in 2022, China has accelerated efforts to reduce its dependence on US semiconductor technology.
This shift toward technological independence has benefited domestic companies like Huawei, which has positioned its Ascend chips as a central alternative to US-made hardware.
