AI Trends

Alibaba Uses In-House Chips to Train Next-Generation Models as AI Product Annualized Revenue Exceeds $10 Billion

WallstreetcnUpdated Nov 19 20:50

AI summary

Alibaba said Qwen has used domestically produced chips from T-Head to train its next-generation models, while annualized revenue from AI-related products has surpassed $10 billion. AI-related product revenue reached RMB 17.9 billion this quarter, accounting for 49% of Alibaba Cloud’s external recurring revenue. The company also plans to begin mass production of its Zhenwu V900 chip in the first quarter of 2027. The newly released Qwen4 Max has 2.4 trillion backbone parameters, with a focus on improving agentic task performance and coding capabilities.

Why read this

SelectedAI score 81

The earnings call disclosed Alibaba’s AI revenue, chip roadmap, and model progress, offering insight into its full-stack AI commercialization strategy.

Topics

Tags

Body · AI translation

On a conference call held Thursday evening, Alibaba’s CEO said the company is using its self-developed chips to train its next-generation models, while annualized revenue from AI-related products has exceeded $10 billion.

Conference call transcript:

This quarter, Alibaba Group’s revenue growth accelerated and profitability continued to improve. Total group revenue grew 10% year over year, driven by the AI, cloud, and computing services segment, where external revenue growth accelerated to 64%. Meanwhile, recurring revenue from external customers grew 67% year over year. Adjusted EBITDA margin rose to 14.5%. Revenue from AI-related products recorded triple-digit growth for the 13th consecutive quarter, with annualized revenue surpassing $10 billion. Profitability improved significantly across our e-commerce businesses. In quick commerce, we maintained market share while substantially narrowing losses.

Rapid improvements in AI model capabilities are accelerating the adoption of AI applications across industries and driving demand for training and inference. The large-scale deployment of AI agents is further boosting demand for computing power. Taken together, these factors mean that global demand for computing power will continue to exceed supply over the next several years. This robust demand provides a solid foundation for the large-scale commercialization of AI.

First, revenue and profit growth in AI and cloud are both accelerating. This quarter, external customer revenue from AI, cloud, and computing services grew 64% year over year, marking the 10th consecutive quarter of accelerating growth. Adjusted EBITDA increased 121% year over year, while EBITDA margin expanded to 14.5%. This performance was jointly driven by growth in computing, storage, MaaS, and AI applications, as well as synergies. AI-related products generated revenue of RMB17.9 billion this quarter, accounting for 49% of Alibaba Cloud’s external recurring revenue. AI has now become a core revenue pillar of our cloud business. Annualized revenue from AI-related products surpassed $10 billion. This strong revenue growth reflects broad customer demand for our “AI + Cloud” services. In addition to training and inference demand from AI laboratories, we continue to see strong demand from customers across industries in China, including large state-owned enterprises and companies in intelligent manufacturing, automobiles, embodied intelligence, and biotechnology.

Second, we continue to expand the scale of T-Head’s AI chips and strengthen our supply chain capabilities. Let me begin with chips: T-Head has established a full-stack portfolio of self-developed chips covering GPUs, CPUs, storage, and networking chips. Since their launch in August, shipments of Super Node instances based on the Zhenwu M890 have continued to ramp up and have now achieved large-scale commercial deployment. We expect to begin mass production of the next-generation Zhenwu V900 chip in the first quarter of 2027. The Zhenwu V900 delivers three times the computing performance of the M890. Next-generation Super Nodes based on the Zhenwu V900 support large-scale network interconnection and can handle workloads such as the training and inference of ultra-large-scale models. These Super Nodes will become a cornerstone of Alibaba Cloud’s AI infrastructure. Zhenwu V900 Super Node clusters will achieve large-scale commercial deployment in the second quarter of 2027. T-Head is accelerating its three-generation GPU roadmap. We have fully rolled out the Zhenwu M890 this year; in 2027, we plan to expand the supply chain and move the V900 into large-scale mass production, followed by the J900 in 2028. Infrastructure powered by domestically produced chips that we developed in-house is becoming the foundation of Alibaba’s full-stack AI capabilities. It will increasingly support model training, inference, and agent workloads for Alibaba Group and Alibaba Cloud customers, underpinning the growth of the group’s AI revenue. Qwen has used T-Head’s domestically produced chips to train its next-generation models, and Model Studio is also providing model inference services based on domestically produced chips. Our full-stack AI infrastructure will give us a clear advantage in this domestic computing-power supply chain. Deep integration between our self-developed chips and models will continue to reduce the unit cost of computing power and help us achieve greater efficiency in AI application development. Taking all these factors into account, our target for calendar year 2030 remains unchanged: $100 billion in external cloud revenue with a margin exceeding 20%.

Third, we continue to advance our models and AI agent applications. Our newly released foundation model, Qwen4 Max, has 2.4 trillion backbone parameters. We continue to enhance its key capabilities, including agentic tasks and coding, areas in which it now ranks among the world’s leading models. Leading partners have also recognized Qwen’s capabilities. Apple Intelligence, which is deeply integrated with Qwen, will launch for users in China in December. On the enterprise side, Qwen Work was released in August and quickly became a leading enterprise-grade AI agent. We recently launched an e-commerce edition of Qwen Work, and the proportion of paying e-commerce users continues to rise. On the consumer side, the Qwen App is becoming a personalized assistant that helps users get things done and currently offers services covering more than 20 everyday scenarios.

Fourth, the profitability of our e-commerce business improved substantially, with both EBITA and margin improving in China e-commerce. This quarter, our e-commerce business delivered solid revenue growth and significantly higher profit. Adjusted EBITA and margin in China e-commerce also improved significantly, demonstrating its stable profitability. Our quick-commerce business maintained its market share, losses continued to narrow, and unit economics improved sequentially. Two quarters ago, we began to see our AI investments cross the commercialization inflection point. Last quarter, growth accelerated and margins expanded. This quarter, these investments are translating into commercial results at a faster pace, with increasingly clear returns. Our years of investment in computing power and cloud infrastructure have given us full-stack product advantages and economies of scale, which are becoming key competitive advantages for Alibaba in the AI industry. They enable us to respond flexibly to industry changes and deliver AI products at scale to meet customers’ rising demand for AI. Looking ahead, we will continue to invest in AI infrastructure to extend our lead and strengthen our long-term competitiveness.

I will now turn it over to Toby to discuss our financial performance. Thank you.

Updating... Risk Warning and Disclaimer

Markets involve risks, and investments should be made with caution. This article does not constitute personal investment advice, nor does it take into account any individual user’s specific investment objectives, financial situation, or needs. Users should consider whether any opinions, views, or conclusions in this article are appropriate for their particular circumstances. Any investment made on this basis is at the investor’s own risk.

Source

Wallstreetcn