Alibaba's Qwen 3.8 Max Runs on All-Domestic Zhenwu M890 Chips, Reports Claim
bookwormengr · x · 2026-09-22
According to a post, Alibaba's new 2.4T-parameter Qwen 3.8 Max model reportedly runs inference entirely on domestic infrastructure — the first time China's homegrown AI stacks officially handle frontier-scale MoE workloads.
- Chips: proprietary Zhenwu M890 with 144GB memory and 800GB/s interconnect bandwidth
- Clusters: Panjiu AL128 supernodes with 128-chip scale-up
- Interconnect: proprietary ALink protocol (NVLink-like), with hints of adopting NPO to scale beyond 128 chips
- The post also frames the domestic stack landscape: DeepSeek+Huawei Ascend vs Moonshot+Alibaba (which holds a 30%+ stake in Moonshot), arguing Zhenwu shouldn't be overlooked despite Ascend's notoriety.
More from Infra
- H Company trains computer-use agents on SkyPilot: thousands of sub-second sandboxes — skypilot_org · 2026-09-23
- MiniMax H3 video gen runs locally on M5 Ultra: 768p in ~2m22s with optimizations — bakawolf123 · 2026-09-22
- Cisco: Agentic AI to Drive 9X Enterprise Traffic Growth by 2035 vs 2.5X Without — Beth_Kindig · 2026-09-22
- Cloudflare ships Vary support in Cache Rules to tame HTTP's 'ugliest' header — threepointone · 2026-09-22
- Skipping failed pipeline stages during training: near-baseline loss at 1% failure rate — covenant_ai · 2026-09-22
- Fluidstack breaks ground on $4B Texas data center campus planned for up to 1.5 GW — MxMnr · 2026-09-22