Google’s newly released Gemini 3 Flash model is now officially available. According to detailed discussions from the Linux.do community, Google One Pro members can directly access this model through the cli2api interface by simply adding the model ID models/gemini-3-flash-preview. Tests show that everything works smoothly when calling the model on the CherryStudio platform, with no technical issues encountered. However, it’s worth noting that the model comes with a thinking mode enabled by default, and CherryStudio may not have been updated to support this feature yet, so developers need to selectively enable the thinking mode. This release not only demonstrates Google’s leadership in AI technology but also provides developers with more efficient tools for integrating and applying AI models. Community members suggest reading the full discussion for more in-depth information and conversation.
