[Comment to Win Points] MiniMax and Zhipu AI are about to release their HK-listed financial reports! How should we assess the commercial "quality" of these two LLM giants?

The AI sector has never lacked hype. In the past, the market loved discussing parameter scale, multimodal capabilities, and open-source ecosystems. But now, more and more investors are turning their attention to a more fundamental question: Can large language models actually make money? Can technological leadership truly translate into user scale and enterprise clients?

$MINIMAX-W (00100.HK)$ and$Z.AI (02513.HK)$ These are essentially the two most representative companies in the topic of LLM commercialization. MiniMax will announce its interim results for 2026 on August 26, 2026,Institutional expectationsMiniMaxReported revenue of $120 million for the first half of fiscal year 2026Zhipu AI will announce its interim results for 2026 on August 31, 2026,Institutional expectationsZhipu AIReported revenue of $1.808 billion for the first half of fiscal year 2026.As key players in the HK-listed LLM camp, their continuous moves in product and technology strategies have also attracted significant market attention. For investors, what may be truly worth watching is not the short-term figures on the report card, but whether the long-term narratives of both companies still hold water.

MiniMax: General-purpose LLM + AI-native Products + Open Platform

First, can the multimodal foundation model continue to deliver differentiated experiences?

MINIMAX has independently developed a series of general-purpose LLMs for text and multimodal applications, and has launched AI-native products globally, along with an open platform for enterprises and developers. The market's key focus lies in whether multimodal capabilities can create perceptible experience advantages for users at the product level, and whether model iterations can directly reflect in user activity and retention.

Second, can the global user base be converted into tangible commercial returns?

Official data shows that MINIMAX's proprietary models and AI products have cumulatively served over 300 million individual users from more than 200 countries and regions, as well as millions of enterprise clients and developers from over 100 countries. The key question raised by this data is not 'whether there are many users,' but rather whether individual users are willing to pay continuously or use the service with high frequency, and whether enterprise clients are truly engaging deeply and retaining long-term.

Third, can the open platform become a moat for the B2B sector?

For enterprises and developers, MINIMAX's strategy is to productize, interface, and solutionize its model capabilities, thereby lowering the barrier to entry. What the market needs to observe more closely is the depth of usage within the developer ecosystem, the willingness of enterprise clients to renew contracts, and whether the open platform can evolve from merely 'providing tools' to becoming a 'business partner.'

Zhipu AI: The GLM matrix continues to iterate, with ecosystem development and commercialization proceeding in parallel under the AGI roadmap.

First, can the model technology maintain a stable lead?

Zhipu's GLM series of models has formed a complete iteration path from GLM-130B and ChatGLM to GLM-4, GLM-4.5/4.6/4.7, and then to GLM-5/5.1, achieving continuous breakthroughs in reasoning, coding, and agents. The core focus of the market is whether technological leadership can be sustained and whether the pace of model iteration can support subsequent commercial deployment.

Second, can open-source initiatives and the developer ecosystem be converted into commercial momentum?

Zhipu entered the open-source large model space relatively early, significantly driving the developer ecosystem. However, the long-term value of an open-source ecosystem depends not just on download counts, but more importantly on the conversion efficiency from 'trying it out' to 'becoming indispensable.'

Third, is the governance and strategic arrangement clear for the transition from a laboratory setting to a long-term commercial entity?

As Zhipu AI moves from its Tsinghua University lab origins to a listing on the HKEX, the alignment between its subsequent technical roadmap and industrialization pace is a key dimension for the market to assess its long-term value.

So what should we really be watching this time?

This time, the market's focus is not just on what new products MiniMax and Zhipu AI are launching individually, but rather on whether the two leading large-model players in the Hong Kong stock market can establish their commercial closed loops as scheduled.

✅ MiniMax: Quality of C-end user base, stickiness of enterprise clients on its open platform, and efficiency in multimodal implementation.

✅ Zhipu AI: Leadership in model technology, quality of its open-source ecosystem, and long-term industrialization pace.

✅ Shared Question: How far have large models progressed in shifting from a "tech narrative" to a "commercial closed loop"?

Now it’s up to fellow investors’ judgment! Come to the comments section to share your opinion and join the discussion to earn reward points! 👇👇

🏆 Event: Bull Circle Opinion Leader ✨

MiniMax and Zhipu AI, the twin giants of large models in the Hong Kong stock market, are competing on the same stage. Which one are you more bullish on?

Share your thoughts in the comments below:

✅ At least 30 words ✅ Original viewpoint ✅ Complies with community guidelines

Eligible participants will share 20,000 points!

Note: All activities above will end at 16:00 (Beijing Time) on August 26; rewards will be distributed uniformly after the conclusion of this earnings season.

Click to reserve:

MiniMax 2026 Interim Results Live Stream

‘Earnings Express’ has been fully upgraded—with three key features to help you get ahead during earnings season!

Futubull AI Get instant answers to your questions, use intelligent stock selection to target promising upward momentum, and analyze your holdings to seize opportunities and avoid risks!

235K Views