China’s AI Lab Zhipu Weighs Custom Chip As Demand for its GLM Model Soars

2026-07-07 · Source: The Information · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Emerging Technologies & Innovation · Depth: Fundamental Awareness, quick

Summary

Zhipu AI, the Chinese AI lab behind the highly regarded GLM series of open-source AI models, is weighing the design of its own custom AI chip. This consideration is driven by surging demand for its models and the increasing constraint on computing resources due to U.S. export controls. The Beijing-based company, a leading AI lab in China, has made preliminary inquiries with several Chinese chip design houses regarding the possibility of developing a bespoke AI processor. This chip would be specifically optimized for efficiently running Zhipu's GLM models. While discussions are still at an early stage, and no designer has been selected, this initiative highlights the strategic response to geopolitical and market pressures impacting AI development in China.

Key takeaway

For Directors of AI/ML managing large model deployments, Zhipu AI's move signals a critical shift towards vertical integration in response to supply chain vulnerabilities. You should evaluate the long-term viability of relying solely on general-purpose hardware, especially if your models have unique computational demands or if geopolitical risks affect your hardware access. Consider exploring custom silicon solutions or diversifying your hardware procurement strategy to ensure operational continuity and optimize performance for your specific AI workloads.

Key insights

Chinese AI labs are exploring custom chips to mitigate U.S. export controls and meet soaring demand for their models.

Principles

• Geopolitical factors drive hardware innovation.
• Custom hardware optimizes model performance.
• Supply chain resilience is critical for AI.

In practice

• Investigate custom silicon for specific model architectures.
• Diversify hardware sourcing strategies.
• Assess geopolitical risks in AI supply chains.

Topics

• Zhipu AI
• GLM Models
• AI Chips
• Custom Silicon
• U.S. Export Controls
• AI Hardware

Best for: Investor, CTO, VP of Engineering/Data, Director of AI/ML, AI Hardware Engineer, Tech Journalist

Related on AIssential

• OpenAI pulls ahead on custom chips — Global stability is increasingly influenced by the interplay of advanced AI infrastructure and complex geopolitical realignments.
• Meituan's LongCat-2.0 shows China can train massive AI models without Nvidia — China has demonstrated the ability to train trillion-parameter AI models using only domestic hardware, challenging US export controls.
• OpenAI Unveils First Custom AI Chip With Broadcom — Custom AI hardware and strategic capital are critical for scaling AI infrastructure and managing costs.
• Why everyone from OpenAI to SpaceX is building their own chips (and turning up the heat on Nvidia) — Major tech firms are developing custom AI chips to reduce supplier dependence and optimize performance.
• How the Global AI Race Has Changed — The global AI landscape is rapidly diversifying with new powers and technological fronts emerging beyond the traditional US-China rivalry.
• Huawei Ascend Production Ramp: Die Banks, TSMC Continued Production, HBM is The Bottleneck — China's drive for AI compute self-sufficiency is bottlenecked by HBM supply, despite significant domestic chip production efforts.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by The Information.