On the main branch, subh05sus updated the AI integration code by replacing all instances of the “gemini-2.5-flash-lite” model with “gemini-2.5-flash” and standardized string quoting across numerous files. This ensures the project now consistently uses the newer Gemini AI model throughout.