appConfigEntityId

bodyType

customContents

entityId

entityUuid

isVideo

articlePageQuery

article

articleSchemaContent

content

hooksArticlePageAdvertisementConfigArticle

hooksArticlePageFloatingButtonConfigArticle

appConfig

filter

json

liveArticle

Richard Chen

Rising cost efficiency and rapidly advancing capabilities from players like DeepSeek and MiniMax have prompted Goldman Sachs to raise its run-rate revenue forecast for China’s artificial intelligence model market by 30 per cent to US$13 billion.
In a research note published on Monday, the US investment bank boosted its year-end annualised recurring revenue (ARR) projection for mainland Chinese AI models from US$10 billion, citing aggressive price cuts, technical breakthroughs and accelerating corporate adoption.
Goldman Sachs lifted projected year-end ARRs of Hong Kong-listed Zhipu AI and MiniMax to US$2.5 billion and US$1 billion, respectively.
“We expect competition for the best performance-to-price balance to intensify among Chinese AI models,” the report said, noting that recent releases from domestic pioneers had reached new frontiers in performance per dollar.
The heightened competition followed MiniMax’s launch last Friday of its H3 model under an open-weight approach. The multi-modal system – capable of processing text, image, video, audio and music – was priced at just 30 to 50 per cent of incumbent market levels, according to Goldman Sachs.
On the same day, DeepSeek officially launched application programming interface (API) access for its V4 Flash model, achieving front-end coding capabilities that rival Zhipu’s flagship GLM-5.2.
According to Arena AI’s latest leaderboard, GLM-5.2 ranks seventh in front-end coding globally, while DeepSeek V4 Flash closely follows in eighth place.

Meanwhile, e-commerce giant Alibaba Group Holding has officially rolled out its massive 2.4-trillion-parameter Qwen3.8 Max model. In Arena AI’s front-end coding rankings, Qwen3.8 Max is in fourth place, outperforming OpenAI’s top-tier GPT-5.6 Sol in sixth place, while offering lower costs.
Alibaba’s Qwen3.8 Max is priced at US$2 per million input tokens and US$6 per million output tokens. For comparison, OpenAI’s flagship GPT-5.6 Sol is billed at US$5 per million input tokens and US$30 per million output tokens.
OpenAI responded last Friday with a counter-offensive, announcing an 80 per cent price drop for its lightweight GPT-5.6 Luna model, slashing API fees to 20 US cents per million input tokens, alongside a 20 per cent reduction for its mid-tier GPT-5.6 Terra.
Alibaba owns the South China Morning Post.
Driven by these cost-effective offerings, Chinese open-weight models now commanded nearly 70 per cent of global token consumption on developer platforms like OpenRouter, according to Goldman’s tracking.
The competitive arena is also expanding to workspace agentic tools. Tencent Holdings’ WorkBuddy topped China’s desktop AI agent rankings in June, capturing a 34 per cent market share by monthly visits, followed by ByteDance’s TRAE at 21 per cent, according to the bank’s research.
To monetise, Chinese developers were expected to increasingly shift towards “community licences” for open-weight models, requiring commercial users to pay fees.
At the lower end of the market, Goldman expected API pricing and gross margins to remain under pressure through the second half of the year, with well-funded model makers using post-fundraising cash to subsidise aggressive pricing.
However, the bank warned of impending geopolitical constraints, noting reports that Chinese authorities were considering restrictions on foreign downloads of model weights and overseas transfers of training data.


Huawei unveils latest tech to boost AI power in push to break China’s Nvidia reliance

Hong Kong’s 5-year plan, policy continuity offer investors more stability, experts say

Huawei accused of stealing robot tech to get hands on T-Mobile’s trade secrets

Hong Kong aims to raise innovation spending to 3% of GDP by about 2030

Artificial intelligence (AI)

China’s AI revenue projected to reach US$13b on breakthroughs, adoption: Goldman Sachs

Goldman Sachs boosted its year-end annualised recurring revenue projection for mainland Chinese AI models to US$13 billion from US$10 billion. Photo: Shutterstock

The MiniMax booth at the World AI Conference in Shanghai, China, on July 17. Photo: AP

China’s Moonshot AI aims for US$50b round as year-end Hong Kong IPO targeted: sources

Kimi K3 has outperformed leading US systems from Anthropic and OpenAI in certain capabilities, according to Moonshot. Photo: Reuters

China tightens chip-design protection and clarifies rights to spur innovation

People learn about a reconfigurable AI chip at the permanent exhibition at the Zhongguancun Exhibition Center in Beijing. Photo: Xinhua

China’s MiniMax curbs overseas access to new AI video model over copyright disputes

MiniMax says it will review deployment scenarios and compliance controls before it can authorise use of the H3 model in some major markets. Photo: Shutterstock

The AI talent war: tech giants court researchers years ahead of graduation

Students pose for graduation photos on the campus of Zhejiang University of Technology, June 2, 2026. Photo: VCG via Getty Images

College students at the Large-scale Recruitment Fair for Science and Engineering Universities in Harbin, China, March 24, 2026. Photo: Getty Images

Tech

Big Tech

Tech Trends

Tech War

Business

China Business

Companies

News

China Future Tech

Bank forecasts a sharp revenue surge after Chinese AI models slash costs, boost capabilities and capture global token share


Bank forecasts a sharp revenue surge after Chinese AI models slash costs, boost capabilities and capture global token share.


China’s AI revenue projected to reach US$13b on breakthroughs, adoption: Goldman Sachs

.css-1c6uqr6{color:inherit;font-weight:inherit;font-size:inherit;font-family:inherit;line-height:inherit;overflow-wrap:break-word;}Bank forecasts a sharp revenue surge after Chinese AI models slash costs, boost capabilities and capture global token share

Bank forecasts a sharp revenue surge after Chinese AI models slash costs, boost capabilities and capture global token share