Shares in Chinese AI company Zhipu AI surged 33%, according to CNBC, after the firm open-sourced its latest language model and a major Wall Street bank placed a higher-conviction bet on the company.
The catalyst on the product side was GLM-5.2, launched on June 13, 2026. According to MarkTechPost, the model ships with a usable one-million-token context window — meaning it can process roughly the equivalent of several full-length novels in a single conversation. It also offers two reasoning modes, labeled High and Max effort, letting developers trade speed for thoroughness depending on the task. The model integrates with popular AI coding tools including Claude Code and Cline through an Anthropic-compatible endpoint. Notably, no performance benchmarks were published at launch, an unusual move in an industry obsessed with leaderboards. The model is released under an MIT open-source license, making it freely available for commercial use.
On the financial side, JPMorgan sharply hiked its price target for Zhipu AI shares, according to Investing.com, lending the rally an institutional stamp of approval. CNBC framed the broader move as Wall Street raising its bets on Chinese AI companies, partly in response to restrictions placed on Anthropic — a signal that investors are hunting for alternatives outside the U.S.
Not every Chinese AI name shared in the gains. According to Nikkei Asia, rival Minimax stumbled as China's AI stocks diverged, suggesting investors are picking winners rather than buying the sector wholesale.
The episode illustrates how quickly open-source releases can move markets — and how U.S. policy decisions around AI exports are reshaping which companies global investors favor.