Key Highlights
- The Zhenwu M890 AI accelerator delivers triple the computing power of the previous 810E model, designed specifically for agentic AI applications.
- Developed by Alibaba’s T-Head semiconductor division, the chip is now accessible to enterprise clients in China through Alibaba Cloud.
- The company revealed a strategic chip development plan featuring the V900 (launching Q3 2027) and J900 (arriving Q3 2028), each targeting approximately 3x performance improvements.
- The company simultaneously launched Qwen 3.7-Max, an enhanced large language model optimized for sophisticated coding tasks and extended agent operations lasting up to 35 hours.
- More than 560,000 Zhenwu processors have been delivered to over 400 clients spanning 20 different sectors.
On Wednesday, Alibaba introduced a domestically-produced AI accelerator alongside an enhanced version of its primary large language model, reinforcing its strategy to establish a Chinese AI infrastructure independent of Nvidia technology.
The freshly announced processor, designated Zhenwu M890, achieves performance levels three times higher than the earlier Zhenwu 810E model. The technology was presented at the Alibaba Cloud Summit held in Hangzhou and is immediately accessible to Chinese business clients via Alibaba Cloud’s Bailian service platform.
Shares of Alibaba (BABA) declined 0.60% during the announcement period.
Alibaba Group Holding Limited, BABA
T-Head, Alibaba’s chip design subsidiary, engineered the M890 with 144GB of GPU memory capacity—a substantial increase from the 96GB featured in earlier versions—enabling the processor to manage significantly larger datasets.
A key distinction from previous iterations is that the M890 supports both model training and inference operations, whereas earlier chips focused exclusively on inference. This dual capability positions it as a comprehensive solution for organizations developing and deploying AI systems.
The processor is specifically engineered for agentic AI applications—intelligent systems capable of executing intricate, multi-stage operations with minimal human oversight. Such computational demands require models to maintain extended contextual memory and coordinate dynamically with other systems.
According to T-Head, cumulative shipments of Zhenwu processors have surpassed 560,000 units, serving more than 400 external clients across 20 sectors including automobile manufacturing and financial services.
Long-Term Development Strategy Revealed
The announcement extended beyond a single product release. Alibaba presented a comprehensive chip development timeline featuring the V900, scheduled for Q3 2027 release, with the J900 following in Q3 2028.
The V900 is anticipated to achieve approximately another 3x performance enhancement compared to the M890. This consistent development pace demonstrates that Alibaba views semiconductor innovation as a fundamental infrastructure investment rather than an isolated initiative.
Complementing the M890 launch, Alibaba introduced the Panjiu AL128, a server architecture that integrates 128 of the new accelerators within a single rack configuration.
Enhanced Language Model Released Simultaneously
Alibaba concurrently released Qwen 3.7-Max, the newest iteration of its leading language model. This model is specifically designed for complex programming tasks and extended autonomous agent operations, maintaining consistent performance for periods reaching 35 hours.
Both the Qwen enhancement and chip introduction were announced together during the Hangzhou summit, presenting them as complementary components of a unified strategy focused on agentic AI capabilities.
This development unfolds as the United States maintains restrictions on exports of Nvidia’s cutting-edge processors to Chinese purchasers. Alibaba previously ranked among Nvidia’s largest Chinese clients, and these export limitations have catalyzed the transition toward domestically manufactured alternatives.
Additional Chinese semiconductor manufacturers such as Biren Technology and MetaX have recently completed public listings on mainland China and Hong Kong stock exchanges, indicating broader momentum in the domestic chip industry.
Alibaba has previously committed more than 380 billion yuan (approximately $53 billion) toward cloud computing and AI infrastructure investments spanning three years—representing the company’s most substantial investment in this domain.
During a recent earnings discussion, CEO Eddie Wu indicated that AI-driven product revenue is projected to emerge as the principal growth catalyst for the cloud division. The organization has characterized its proprietary chips as delivering “the highest value for money compute power,” with potential to enhance profit margins.
Get 3 Free Stock Ebooks
Discover top-performing stocks in AI, Crypto, and Technology with expert analysis.
- Top 10 AI Stocks - Leading AI companies
- Top 10 Crypto Stocks - Blockchain leaders
- Top 10 Tech Stocks - Tech giants
