In the Linux.do community, users are actively discussing performance changes in the Gemini web version. Since the 3 Flash model update, multiple users have reported that the Gemini web interface has become difficult to use. The fast version is considered severely underpowered, while the pro version fails to meet professional standards. Specific issues include poor adherence to thinking instructions, inability to complete simple tasks even after repeated communication, and the model cutting corners in its reasoning process. Users have expressed that their recent experience has been extremely frustrating, to the point of wanting to vent their anger. This topic spans 12 posts with 8 participants, reflecting the challenges AI models face in practical applications. For readers following cutting-edge AI technology, this feedback provides a valuable user perspective, highlighting the importance of model performance optimization.
