Alibaba surges over 6%! Why is the Hong Kong stock market experiencing a short squeeze?

On April 8th, the A-share and Hong Kong stock markets all surged collectively, and the Hong Kong tech sector in particular swept away pessimism, staging a “short squeeze” rally. Alibaba, with a total market value of over 2.4 trillion yuan, saw its intraday gains at one point expand to 7%, and ETFs such as the Hang Seng Internet ETF Huaxia (513330.SH) also followed with a sharp rise.

Industry insiders analyze that the reason Hong Kong tech stocks staged a short squeeze rally is, on the one hand, that the marginal easing of geopolitical crises prompted funds to flow first into “valuation valleys” that are sensitive to liquidity. On the other hand, domestic economic data for January and February came in above expectations; starting in Q2, PPI turned positive, and the macro economy is expected to face a recovery inflection point. Finally, the proportion of short sellers in Hong Kong stocks is at a high level; with global stock market volatility clearly rising, the risk-reward ratio of leveraged long-and-short trading has fallen, and some short sellers have gradually closed their positions.

On the news front, AI business updates have been coming in frequently as the internet giants led by Alibaba, boosting the valuation of their technology growth.

On April 7th, OpenClaw founder Peter Steinberger announced: The next-generation OpenClaw product will incorporate video generation functionality, and will be the first to provide official support for Alibaba’s Qwen large model.

Last week, according to OpenRouter, China’s AI large model token calls reached 12.96 trillion Token, up 31.48% from the previous week, surpassing the United States for five consecutive weeks, with the gap continuing to widen. Among them, Alibaba holds two seats on the TOP5 list: Qwen3.6Plus (free) topped the weekly chart with 4.6 trillion Token in weekly call volume; in addition, Qwen3.6PlusPreview ranked third on the weekly list.

On April 7th, Alibaba’s AI assistant Qianwen received its latest upgrade of “In-depth Research” professional capabilities, adding new modules such as financial analysis, and integrating real-time data from 13,000 stocks and around one million listed-company financial reports. On April 3rd, Alibaba released a newly upgraded Wan2.7-Video video generation model, covering four major models: text-to-video, image-to-video, reference-based video, and video editing. This expands AI capability from single-material generation to the full end-to-end creative workflow. On the same day, Code Arena announced its latest round of rankings: Alibaba’s Qwen 3.6-Plus ranked second on the global list, surpassing international giants such as OpenAI, Google, and xAI, becoming the highest-ranked Chinese large model on that list. On April 2nd, Alibaba released its latest next-generation large language model—Qwen3.6-Plus—with native multimodal understanding and reasoning capabilities, and it performs especially well in code generation and Agent capabilities. It is reported that the more powerful flagship model Qwen3.6-Max will also be released soon.

Pay attention to the Hang Seng Internet ETF Huaxia (513330.SH), which passively tracks the Hang Seng Internet Technology Industry Index. Listed in A-shares, it comprehensively covers a basket of major internet platform giants (Alibaba’s share is 15%; Tencent Holdings + Alibaba + Meituan + NetEase + JD.com + Baidu together account for over 62%). At the same time, its unit price is only a few dozen to a hundred yuan, and it also comes with an advantage of low fees.

In the mobile internet era, internet giant companies have accumulated advantages in engineering talent, advantages in application entry points, and massive user moats. They have strong capabilities in integrating AI technology and ecosystems, along with clear advantages in deploying to real-world scenarios. In addition to Alibaba and Baidu Group, there are also independent domestic semiconductor chip manufacturers under their umbrella—Pingtouge and Kunlun—whose computing power risks are relatively lower than those of other small model vendors.

Daily Economic News