A new artificial intelligence model developed by Chinese startup Z.ai is attracting widespread attention after industry experts said it delivers capabilities approaching those of leading US models at significantly lower cost.
The model, GLM-5.2, has been described by some technology leaders as a “mini DeepSeek moment”, reigniting debate over whether China is rapidly closing the gap with the United States in frontier AI.
Silicon Valley Takes Notice
GLM-5.2 has gained popularity among developers for its strong coding performance and advanced AI agent capabilities, enabling it to carry out complex tasks with minimal prompting.
The model has climbed developer rankings on platforms such as OpenRouter, outperforming some offerings from Anthropic, while executives including Snowflake CEO Sridhar Ramaswamy and venture capitalist Marc Andreessen have praised its performance.
Former White House AI adviser David Sacks said the model is “right up there” with OpenAI and Anthropic’s latest systems, warning that the US cannot afford policies that slow its own AI companies.
Performance at Lower Cost
According to benchmark rankings, GLM-5.2 is among the world’s top-performing large language models and is especially strong in coding tasks.
Researchers say it operates at roughly one-sixth of the cost of comparable closed-source models such as OpenAI’s GPT series and Anthropic’s Claude models.
Unlike many open-source systems, developers say GLM-5.2 can be deployed with minimal customisation, making it easier for businesses to adopt.
Growing Interest Beyond China
Industry experts say demand for lower-cost AI models has increased as businesses face rising costs when using commercial AI services.
Some analysts also argue that uncertainty surrounding US AI regulation has encouraged developers to explore alternative models.
However, concerns over data security and geopolitical tensions remain major barriers to widespread adoption by US companies, particularly in regulated sectors such as finance and cybersecurity.
Can It Challenge OpenAI?
Experts say GLM-5.2 is unlikely to replace leading US AI models overnight but could become an increasingly attractive option for developers, startups and small businesses seeking lower-cost alternatives.
According to a RAND report, the global market share of Chinese large language models rose sharply following the launch of DeepSeek, particularly in developing countries and nations with close economic ties to Beijing.
While questions remain over enterprise adoption in the West, GLM-5.2 is widely seen as another sign that China’s AI industry is becoming an increasingly competitive force in the global race for artificial intelligence.
(with inputs from Reuters)
