What is GLM-5.2? Another open-source Chinese AI model has Silicon Valley’s attention.

• There is a new Chinese artificial intelligence model that is making a big impact in Silicon Valley.

• It’s called GLM-5.2 and is designed for long-running coding tasks.

• “I am truly impressed,” Vercel CEO Guillermo Rauch wrote to X.

A new AI model from China is generating excitement not seen since DeepSeek’s R1 declared China a serious threat to American chatbot hegemony a year ago.

Silicon Valley’s online echo chamber has been alight in recent days with intrigue surrounding z.AI’s new open-source model. Called GLM 5.2, it is a broad language model designed to handle long coding tasks and mediated workflows.

The company says it’s working on a 1 million token content window, which would put it in the same league as Anthropic’s. Claude Opus 4.8 And OpenAI’s GPT 5.5.

“I’m really impressed, almost shocked, at how good @zai_org’s GLM-5.2 is at encoding. This changes things.” Guillermo RauchThe CEO of Vercel, a cloud-based platform for developers, wrote on X:

On social media, investors, founders and technology industry leaders expressed their admiration for the speed and capacity of the new model, which was launched last week.

Matt Velloso, former vice president of Meta, Google DeepMind and Microsoft, said in an X post that he spent all day using GLM-5.2. “The first open model to break the bar as a daily driver,” he wrote. “Things won’t be the same.”

Like DeepSeek, GLM-5.2 is open source; This means anyone can download it and run it on their own on their own systems, modifying it as they see fit. Most American boundary models, such as those from OpenAI and Anthropic, are closed.

In closed models, the consumer is dependent on the provider. This is good for the provider because they can capture more of the value required for companies that spend billions of dollars on AI infrastructure and where investors are worried about revenue growth.

But if an open model is just as good or better, it can easily capture a larger share of the market.

The USA and China have been in a competition over the superiority of artificial intelligence in recent years. Washington is trying to maintain its superiority through chip restrictions and access controlsChinese companies are trying to move forward with cheaper and increasingly capable open source models.

Anthropic recently warned in a report that China is moving closer to the United States through looser chip controls, saying, “distillation attacksAnthropic said the United States and its allies still have a chance to “be 12-24 months ahead in border capabilities,” but warned that “the window of opportunity to provide that lead will not remain open for long.”

China first gave Silicon Valley a wake-up call in January last year. DeepSeek releases R1A low-cost reasoning model that rivals OpenAI’s o1. At the time, investors were questioning whether Silicon Valley’s AI leadership was as safe as it seemed.

GLM-5.2 also asks the same question while surfing the internet.

If you like this story, don’t forget to follow Business Content on Yahoo.