Zhenwu V900

Alibaba's AI training and inference chip with 216GB of memory, built to support 500,000-chip superclusters for future Qwen models.

Alibaba · AI hardware and silicon · infrastructure

Published in Feed

Overview

Alibaba's chip subsidiary T-Head unveiled the Zhenwu V900 AI accelerator at the Apsara Conference 2026 in Hangzhou on September 22, 2026. Alibaba CEO Eddie Wu called it the most powerful AI chip in China, claiming three times the performance of the prior-generation M890. It has 216GB of GPU memory, 1,200GB/s of inter-chip bandwidth, and native FP8/FP4 low-precision compute support, and is designed to scale into clusters of up to 500,000 accelerator cards to train future Qwen models in the 5-10 trillion parameter range. Mass production is targeted for Q1 2027, alongside a plan to expand Alibaba Cloud's global data center capacity past 20 gigawatts by 2032.

This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.

Key features

  • 216GB GPU memory, 1,200GB/s inter-chip bandwidth
  • Native FP8 and FP4 low-precision compute support
  • Claimed 3x performance of Alibaba's prior-generation M890 chip
  • Scales to superclusters of up to 500,000 accelerator cards
  • Built to train Qwen models targeting 5-10 trillion parameters

Use cases

  • Large-scale AI model training within Alibaba Cloud infrastructure
  • High-bandwidth inference serving for frontier-scale models
  • China-domestic alternative to export-restricted Nvidia accelerators

Pricing

unknown; not yet commercially available

Pricing reflects the source record and may change. Confirm current terms with the provider.

Dates and sources

Released
Not listed
Record updated
24 Sept 2026
Record last checked
24 Sept 2026

Explore the Feed

Other entries filed under ai hardware and silicon; inclusion does not imply equivalent features.

  • Azure Maia 200 and Maia 300

    Microsoft says its Maia 200 AI accelerator is in production serving Copilot, and unveiled Maia 300 for 2027 deployment.

  • NVIDIA Vera Rubin NVL144

    NVIDIA's next-generation rack-scale AI system pairing Vera CPUs with Rubin GPUs, detailed with a shipping timeline pulled forward to compete with AMD's newly announced Helion rack.

  • AMD Helion

    AMD's open rack-scale blueprint for agentic AI infrastructure, with CPU-only configurations up to 32,768 EPYC cores and an integrated model-call token router.

Browse the Feed