Key Points
- On December 31, 2024, Alibaba open-sourced its next-generation image generation model, Qwen-Image-2512, leading to significant surges in Chinese tech stocks like BlueFocus (Lan Se Guang Biao 蓝色光标) which hit a 20% price limit.
- Alibaba (A Li Ba Ba 阿里巴巴) is making a massive investment of ¥380 billion RMB ($53.2 billion USD) over three years into cloud and AI hardware infrastructure, explicitly stating readiness to increase funding further.
- The global AI landscape is rapidly shifting, with Meta acquiring Manus (which processed over 147 trillion tokens), and China’s Zhipu AI (Zhi Pu 智谱) launching a Hong Kong IPO aiming for an estimated market value exceeding HK$51.1 billion.
- Leveraged “margin” funds have poured over ¥4.3 billion RMB ($602 million USD) into domestic software companies since November, indicating aggressive positioning by institutional investors in AI-adjacent and software plays.
- Brokerages like CITIC Securities (Zhong Xin Zheng Quan 中信证券) project that by 2026, AI models’ logical reasoning capabilities will improve, enabling AI Agents to shift from “cost reduction” to “revenue generation” across various industries.
- BlueFocus: 20% Increase (Daily Limit)
- Leo Group: 10% Increase (Daily Limit)
- Fshld: >10% Increase
- Dot-C United: >10% Increase
The AI applications sector experienced a breakout that had investors scrambling to position themselves.
BlueFocus Intelligent Communications Group (Lan Se Guang Biao 蓝色光标) hit the 20% price limit.
Fshld (Fu Shi Kong Gu 福石控股) and Dot-C United Inc (Yi Dian Tian Xia 易点天下) surged by over 10%.
Leo Group (Li Ou Gu Fen 利欧股份) reached its daily limit, with Prophethub (Xuan Ya Guo Ji 宣亚国际) and Shanghai Hand Enterprise Solutions (Han De Xin Xi 汉得信息) following closely.
The catalyst?
Alibaba’s next-generation AI model just got open-sourced.
Alibaba’s Tongyi Qwen Triggers New Market Catalyst
Fresh catalysts for the AI sector emerged after the market close on December 31.
According to Jiemian News, Alibaba (A Li Ba Ba 阿里巴巴) officially open-sourced its next-generation image generation model: Qwen-Image-2512.
Here’s what makes it significant:
- Significant improvements in skin texture rendering
- Enhanced natural texture restoration
- Advanced complex text rendering
- Generates high-quality images with visible fine details like human hair from simple text prompts
- Supports creation of manga-style PPTs and data infographics
Translation: This isn’t just another AI model drop.
It’s solving real design problems that professionals actually face.
This follows an earlier announcement on December 25, when Alibaba Cloud’s Tongyi Qwen team revealed the open-sourcing of the image editing model Qwen-Image-Edit-2511.
The previous 2509 version had a problem.
Developers noticed slight “drifting”—where subjects or objects moved slightly during editing.
The 2511 version introduced major enhancements to ensure consistency.
That’s the kind of incremental improvement that turns a cool prototype into a tool people actually use.
Find Top Talent on China's Leading Networks
- Post Across China's Job Sites from $299 / role
- Qualified Applicant Bundles
- One Central Candidate Hub
Your First Job Post Use Checkout Code 'Fresh20'
Alibaba’s Massive AI Bet: ¥380 Billion in Infrastructure Over Three Years
Here’s where things get really interesting.
Alibaba (A Li Ba Ba 阿里巴巴) was among the first tech giants to explicitly announce its AI investment scale.
The number?
¥380 billion RMB ($53.2 billion USD) over the next three years into cloud and AI hardware infrastructure.
Let that sink in.
That’s more than their total investment over the previous decade combined.
The intensity of this commitment is already visible in the numbers.
Latest financial reports show:
- Over the past four quarters, Alibaba’s cumulative capital expenditure in “AI + Cloud” infrastructure reached ¥120 billion RMB ($16.8 billion USD)
- During an earnings call in November 2025, Alibaba management stated that the initial ¥380 billion RMB ($53.2 billion USD) strategic investment plan was “conservative”
- They expressed readiness to increase funding further as market demand outstrips supply
Translation: Alibaba isn’t just betting on AI.
They’re betting big, and they’re just getting started.
ExpatInvest China
Grow Your RMB in China:
- Invest Your RMB Locally
- Buy & Sell Online in CN¥
- No Lock-In Periods
- English Service & Data
- Start with Only ¥1,000
Global AI Landscape: Meta Acquires Manus, Zhipu AI Goes Public
While Alibaba is making moves domestically, the global AI landscape is shifting fast.
On December 30, according to China News Service, Manus announced it would be joining Meta.
Here’s what that means for the market:
- Manus will continue to provide products and subscriptions via its app and website
- Maintaining operations in Singapore
- As of early December, Manus had processed over 147 trillion tokens
- Created more than 80 million virtual computers
That’s a significant player getting absorbed into Meta’s ecosystem.
Concurrently, according to Shanghai Securities News, Zhipu AI (Zhi Pu 智谱)—a major player in China’s AI applications space—launched its Hong Kong IPO on December 30.
The details:
- Subscription period runs until January 5, 2026
- Planned listing on the Main Board of the Hong Kong Stock Exchange on January 8, 2026
- Stock code: “2513”
- Plans to issue 37,419,500 H-shares
- Estimated IPO market value exceeding HK$51.1 billion (approximately ¥46.8 billion RMB or $6.55 billion USD)
Major AI players going public.
Established tech giants acquiring emerging AI companies.
The market is consolidating around AI applications at breakneck speed.
Resume Captain
Your AI Career Toolkit:
- AI Resume Optimization
- Custom Cover Letters
- LinkedIn Profile Boost
- Interview Question Prep
- Salary Negotiation Agent
Follow the Money: ¥4.3 Billion in Leveraged Funds Flooding Domestic Software
If you really want to understand where the smart money is going, follow the leverage.
Data from Choice by East Money Information (Dong Fang Cai Fu 东方财富) tells a fascinating story.
Since November, leveraged “margin” funds have net-bought several domestic software concepts.
The top 20 stocks by net margin purchases saw a total inflow exceeding ¥4.3 billion RMB ($602 million USD).
This is serious capital positioning.
Leading the pack:
- Shanghai Hanxin IT with a net purchase of over ¥800 million RMB ($112 million USD)
- Kingsoft Office (Jin Shan Ban Gong 金山办公) with over ¥700 million RMB ($98 million USD)
Then came a long tail of companies receiving net margin purchases ranging from ¥100 million RMB ($14 million USD) to ¥330 million RMB ($46.2 million USD):
- WinSuns (Ying Shi Sheng 赢时胜)
- Hymson (Hai Mu Xing 海目星)
- 360 Security Technology (San Liu Ling 三六零)
- CAS Aerospace (Zhong Ke Xing Tu 中科星图)
- Sifang Luck (Si Fang Jing Chuang 四方精创)
- Orbita (Hang Yu Wei 航宇微)
- Rongke Technology (Rong Ke Ke Ji 荣科科技)
- Aisino (Hang Tian Xin Xi 航天信息)
- Inspur Software (Lang Chao Ruan Jian 浪潮软件)
- Aerospace Software (Hang Tian Ruan Jian 航天软件)
- Kedacom (Su Zhou Ke Da 苏州科达)
- Haige Communications (Hai Ge Tong Xin 海格通信)
What’s the pattern here?
Software, infrastructure, and AI-adjacent companies.
Institutional investors aren’t just watching AI applications from the sidelines.
They’re using leverage to position aggressively.
What Brokerages Actually Think Will Happen in 2026
Multiple brokerages are bullish on AI applications performance in 2026.
And their reasoning is data-backed, not hype-driven.
Kaiyuan Securities (Kai Yuan Zheng Quan 开源证券):
In a research report, they noted that policies are likely to accelerate the penetration of:
- AI search
- AI proofreading
- Vertical-specific agents in government sectors and service industries
- Applications in industrial design and information services
Translation: The real money isn’t in general-purpose AI models anymore.
It’s in specific, practical applications that solve real problems for real industries.
CITIC Securities (Zhong Xin Zheng Quan 中信证券):
They laid out a specific thesis for 2026:
- By 2026, the logical reasoning capabilities of models will improve further
- This supports deployment of AI Agents across various fields
- Enterprise-side AI will shift from “cost reduction” to “revenue generation”
- Key areas with clear potential include: data analysis, code generation, recruitment, sales assistance, and intelligent customer service
- AI revenue contributions for relevant domestic companies are expected to maintain a rapid growth trend in the first half of 2026
That’s a fundamental shift in how companies think about AI.
It’s not about replacing workers or cutting costs anymore.
It’s about AI agents generating entirely new revenue streams.
What This Means for Investors and Founders Right Now
Here’s the actual takeaway from all of this:
We’re watching a market transition in real-time.
- Alibaba is betting ¥380 billion RMB ($53.2 billion USD) on infrastructure because they know AI applications need compute power
- Zhipu AI is going public because investors believe AI applications are revenue-generating businesses now
- Leveraged funds are pouring ¥4.3 billion RMB ($602 million USD) into software plays because they see AI applications integrated into existing enterprise stacks
- Brokerages are projecting its rapid growth in AI applications through 2026 based on concrete use cases, not speculation
If you’re building in the AI applications space, this is validation.
If you’re investing, the smart money is already moving.
If you’re just watching?
Now you know where the actual opportunities are in AI applications.
