TechFlow News, June 16: J.P. Morgan released a research report analyzing China’s large AI model industry. The report notes that as monetization pathways for large models converge toward enterprise workflows, API consumption, coding, and agents, investors are increasingly focusing on models’ practical capabilities, task completion rates, and pricing power. Against the backdrop of AI adoption demand still outpacing inference compute supply, developers’ rapid retreat from premium pricing reflects market skepticism regarding improvements in their models’ capabilities.
Within the sector, the firm holds a relatively positive view of Zhipu AI (02513.HK) and reaffirms its “Overweight” rating for the company. The firm highlights Zhipu AI’s demonstrated “price-maker” behavior—having doubled its API prices this year while sustaining continuous business volume growth—strongly validating market recognition of its large models’ value. The firm raised its revenue forecasts for Zhipu AI for fiscal years 2026–2030 by 26% to 42%, lowered its forecast for adjusted net losses, and significantly increased its target price from HK$950 to HK$1,400. (Jinshi)
