Gemini 3 Pro - A Game Changer!

Has anyone had a productivity lift using Gemini 3 Pro vs Claude or GPT? Over the last few day since changing to Gemini 3 Pro as the lead LLM, the difference has been finally optimal! At a scale that increases productivity per tasks on UI/UX, code etc. that contrasts work from other models on a task by task basis (task x min. = hrs. per task). Conversatons, implementations, logic, maybe even reasoning, has scaled to a point far above that of Claude and GPT. Until now, 10 am PST on 11/24/25. The Gemini 3 Pro model has been reduced to be more similar to Claude and GPT in productivity, length of task and tone. Wondering if Base is layering on code to throttle model intelligence to lengthen task productivity to ensure user time on screen, token use and subscription revenues. If true - shame on you Base44!