Procurement

Post Product

  • Post Supply
  • Manage Supplies

Chinese AI Models Top Global Usage for 24 Consecutive Weeks but Overseas Revenue Lags, Say Experts

   2026-10-14 Yicai / First Financial News20

According to the latest OpenRouter data, weekly invocation volume for Chinese AI large models has exceeded that of the United States for 24 consecutive weeks, remaining firmly in first place globally. However, that traffic growth has not translated into p

According to the latest OpenRouter data, weekly invocation volume for Chinese AI large models has exceeded that of the United States for 24 consecutive weeks, remaining firmly in first place globally. However, that traffic growth has not translated into proportional overseas revenue for Chinese model providers.

The gap is attributed to pricing strategy, ecosystem models and distribution channels. Speaking at the 2026 Forum on China and the Global Economy hosted by Business China in Singapore, Zhang Qian, founding partner of FutureX Capital, told Yicai that the trend of overseas revenue lagging market share expansion is expected to reverse, and that it was "only a matter of time."

Zhang said commercial revenue for closed-source models may decline while open-source model revenue rises. Despite being far from the tens of billions of dollars in revenue run by OpenAI and Anthropic, Chinese model commercial revenue is growing quickly.

She explained that as the capability gap between Chinese open-source models and frontier closed-source models narrows, leading AI application companies are adopting Chinese open-source models as their underlying technical foundation. Harvey, a US legal technology company whose early products relied on OpenAI general-purpose models, launched its first proprietary model, Harvey Tenet, post-trained on Kimi K3, in August this year, citing improved legal scenario performance, lower inference cost and greater technological autonomy. GenSpark chose MiniMax M3 open weights as its base, post-trained with Fireworks AI, and released its first in-house model, Gen-1 Slides, for presentation generation, in September.

Zhang said the trend reflects leading companies' emphasis on technological autonomy, noting that the bigger a company is, the more it values its own AI capability and the less willing it is to build on someone else's ecosystem model. Open-source architectures give enterprises the freedom to customize proprietary models using their own private data.

Price competition is also intensifying. Zhang observed that the high-profit golden period for a leading US model averages only about six months before price cuts begin.

Source: Yicai / First Financial News

 
ReportCollect0Reward 0Comment 0
Disclaimer
• 
This article is an original work by {author}. Reproduction is welcome, but please indicate the original source: {linkurl}. The views expressed in this article are those of the author alone, and the website has not verified the content. Readers are advised to use it for reference only. If the article involves content that violates public morality or laws, it will be deleted immediately upon discovery, and the author shall bear the corresponding responsibilities. In case of copyright or other issues, please contact us in a timely manner.
 
More>Similar News
Recommended images and text
Recommend News
Click to rank

Sign Up

For enterprise

Tel:132-7094-5888

Scan with Phone

Support

Tel:132-7094-5888

Program

Mini Program

Procurement Advisory

Scan with WeChat

WeChat

Business Opportunities

Scan with WeChat

Top