Cross-Sector entity
Of the tracked stories, 2 of 2 also mention GLM-Image, the most common co-covered peer. Coverage clusters in ai-models, which accounts for 1 of those 2, with the remainder spread across 1 other category. Each carries 3 original sources on average.
2 verified stories tracked
Last mentioned: Jun 18, 2026
Entity pulse
Figures are computed live from our source-verified story record — see our methodology for how impact and sentiment are derived.
Of the tracked stories, 2 of 2 also mention GLM-Image, the most common co-covered peer. Coverage clusters in ai-models, which accounts for 1 of those 2, with the remainder spread across 1 other category. Each carries 3 original sources on average. We currently track 2 Cross-Sector stories that mention Ascend 910, all published on June 18, 2026.
Computed from the 2 stories linked to this entity. Beat comparisons are omitted because no baseline was available for this window.
Coverage cohort
Other entities that clear the same relevance threshold in stories also covering Ascend 910. Shared-story counts are live from our verified record — not editorial picks.
As China’s top AI labs begin training models on domestic silicon instead of Nvidia, the shift creates a huge opening for chip startups, model optimization platforms, and training-as-a-service ventures. The first multimodal model fully trained on Huawei Ascend chips signals a decoupling that could reshape the entire AI hardware supply chain.
The first multimodal model trained entirely on Chinese chips — Zhipu’s GLM-Image — proves domestic hardware can handle serious AI workloads, but large language model pre-training remains elusive. The technical hurdles and software ecosystem gaps underscore a multi-year journey toward Nvidia independence.
Source: Xinmei Shen (hk) · Xinmei Shen (cn)
