A Chinese tech giant is rolling out another artificial intelligence model that may give leading American labs a run for their money — but experts say the bigger fight is over who uses it.

Alibaba announced the launch of Qwen4-Max, its most advanced AI model yet, which scored higher than Anthropic’s Claude Fable 5.1 and Moonshot’s Kimi K3.1 Pro in benchmark coding tests, Bloomberg reported Sunday. The company’s announcement follows U.S. policymakers scrambling, after Nvidia CEO Jensen Huang’s Thursday resignation, over how to close export-control loopholes that Chinese companies may have used to access restricted chips.

“Qwen4-Max advances real-world productivity through greater proficiency with professional tools, stronger long-horizon task execution, and higher-quality deliverables,” Alibaba’s Qwen team said in the company’s Sunday announcement. “It can take on a wide range of creative and professional work, spanning 3D modeling, game development, precision manufacturing, finance, law, healthcare, and frontier scientific discovery.”

The new Chinese model currently ranks 6th for coding, above Claude Fable 5.1 and GPT-6.1 Sol, according to Arena.ai’s Code Arena leaderboard. However, Anthropic’s Claude Fable 5.2 ranked no. 1 and OpenAI’s GPT-6 Astra ranked no. 5 for coding, outperforming Qwen4-Max on the same leaderboard at the time of reporting.

Alibaba did not immediately respond to the DCNF’s request for comment.

“Qwen4-Max is clearly a good model, and Alibaba deserves credit for developing a model that delivers strong performance at a relatively low cost, Daniel Castro, president of the Information Technology and Innovation Foundation, told the Daily Caller News Foundation. “Its biggest competitive advantage appears to be efficiency rather than outright capability.”

The model is efficient for longer tasks like coding and legal work, but it is less competitive with American frontier models on tasks involving the creation of visuals or work that reflects real-world jobs, Castro told the DCNF. He considers the model competitive but marginally ahead of its rivals’ models where it leads.

Qwen4-Max is a closed-weight model, unlike open-weight Chinese models such as DeepSeek’s V4, Moonshot’s Kimi models and Z.ai’s GLM-5.3, which can be downloaded for free and modified to fit a user and their system’s needs.

“One interesting aspect of this release is that Qwen4-Max is not an open-weight model, which raises questions about China’s broader strategy of positioning itself as a champion of open-source AI,” Castro told the DCNF.

Castro also said a model like Qwen4-Max may put American businesses at risk if they build on the model.

“With a closed model like Qwen4-Max, businesses are more dependent on the provider’s systems and infrastructure to access its capabilities,” Castro added. “If that means relying on Chinese-based infrastructure or services, it raises additional questions about data security, privacy, and long-term dependency.”

Strong benchmark scores do not explicitly reflect that a company leads the industry, according to Jason Hsu, a senior fellow at the Hudson Institute.

“Americans should take Alibaba’s claims seriously, but not at face value,” Hsu told the DCNF. “What matters is how these models perform in real-world applications, including reliability, security, and scalability.”

Hsu said Chinese companies may not need the best model to win the AI race.

“China doesn’t need to build the world’s most powerful AI model to challenge American leadership,” Hsu told the DCNF. “By offering capable, low-cost open-weight models, Chinese companies can capture global developer adoption and shape the future AI ecosystem.”

“The AI race is no longer just about who builds the best model, but whose technology the world builds on,” Hsu added.

Hsu urged the U.S. to invest in open-weight American models instead of banning Chinese models outright in order to win the AI race.

“America could win the frontier AI race but still lose the global adoption race — and that would have profound economic and geopolitical consequences,” Hsu told the DCNF.


Content created by The Daily Caller News Foundation is available without charge to any eligible news publisher that can provide a large audience. For licensing opportunities of our original content, please contact licensing@dailycallernewsfoundation.org

Dylan Kresak

Share
Published by
Dylan Kresak

Recent Posts

Buffett Gives Away $5.5 Billion in Berkshire Stock and Says Kindness ‘Compounds’ Better Than Anything He Bought

Warren Buffett donated about $5.5 billion in Berkshire Hathaway shares to four family foundations Monday,…

24 minutes ago

United Airlines Hit by a ‘Widespread System Outage’ That Stalls Departures Days After the FAA Ground Stop

United Airlines said Monday it was working to fix a “widespread system outage” affecting departing…

24 minutes ago

Chase Blames a Software Bug, Not a Cyberattack, After Customers at Major Banks Can’t Log In or Send Money

JPMorgan Chase said a software bug in its own systems delayed transactions Monday and may…

48 minutes ago

Nasdaq Sinks 2.5% as Nvidia Leads an AI Rout and Fed Chair Warsh Warns a Bust Could Raise Unemployment

Nvidia led another tech sell-off Monday, dragging the Nasdaq down 2.5% on the same day…

59 minutes ago

Supreme Court Won’t Hear Challenge to Maryland’s ‘Sensitive Places’ Gun Ban, but Thomas Says He Has ‘Serious Doubts’

The justices turned away a challenge to Maryland’s limits on carrying guns in parks, transit,…

3 hours ago

Thanksgiving Flights Cost 17% More and Gas Is on Track for a Holiday Record, as AAA Predicts Fewer Travelers

AAA expects 81.5 million Americans to travel for Thanksgiving, about 121,000 fewer than last year,…

3 hours ago