Story

On air. The free wire runs 15 minutes behind live. See pricing

World, crypto, finance and tech news, scored for impact.

CryptoTechFlowChinese, translatedFiled 2 days ago

Alibaba's Wu Yongming: Qwen has already trained the next-generation model on T-Head chips.

Original headline阿里吴泳铭:千问已在平头哥芯片上训练下一代模型

Open the sourceReported by TechFlow, techflowpost.com, 19 November 2026, by TechFlowDaily

Not investment advice. Muhabbit scores how much a story may matter, not which way to trade.

Alibaba CEO Eddie Wu said on the company’s fiscal 2027 second-quarter earnings call that Qwen has begun training its next-generation models on T-Head (Pingtoutou) chips, with the Bailian platform already serving model inference on domestic silicon. Wu outlined an accelerated three-generation GPU roadmap: Zhenwu M890 is rolling out fully this year, followed by V900 in 2027 and J900 in 2028, with supply chains moving toward mass production. Alibaba’s in-house compute infrastructure is set to progressively handle model training, inference and agentic workloads for the group and Alibaba Cloud customers, underpinning AI revenue growth. The disclosure matters because it signals Alibaba’s push to reduce reliance on foreign accelerators and build a domestic chip stack around Qwen. Next, watch M890 deployment volumes, the V900 timeline and how quickly Alibaba Cloud shifts customer workloads onto its own silicon.

Muhabbit's AI wrote this summary in its own words from TechFlow's report and translated it from Chinese. It can get details wrong, so check the source.

Named in this story

Alibaba (BABA), PingTouGe, Qwen, Wu Yongming, Bailian

A linked name opens that asset's sentiment page.