'Bullish Arrival' Model Mystery Nearly Solved: An Error Points Ox Alpha to Zhipu GLM

Beating AI Newsflash: The identity of Ox Alpha is becoming increasingly difficult to conceal. Community developer Chetaslua sent a deliberately erroneous API request to the model via OpenCode. The backend unexpectedly returned a Java stack trace, which included `paas/v4/chat`, directly matching Zhipu's official API path. He then conducted a comparative experiment. Multiple GLM models hosted by Zhipu consistently returned the same `1214 Incorrect role information` error. However, when the same GLM weights were hosted on DeepInfra, the error format changed. This indicates that it's not just the model that "looks like GLM"—even the backend aligns with Zhipu. Previous model fingerprinting also pointed to GLM. Across 25 text tests, the Token count of Ox Alpha consistently differed from GLM-5.3 by 75 tokens; in four different video tests, the visual Token consumption was completely identical to GLM-5V-Turbo. It can now be highly confirmed that Ox Alpha's backend is powered by Zhipu, and the model itself is highly suspected to belong to the GLM series. However, whether it is a new version of GLM-5.3 or an unreleased new model remains unconfirmed. [Original Article Link]