Google DeepMind has rolled out two new versions of its Gemini chatbot. The company calls them Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking. The launch gives users a clear choice between a speedy conversational assistant and a slower, more careful thinker designed for difficult problems.
The standard Live model handles quick back-and-forth chats, simple answers, and creative tasks in real time. It is built for everyday use where waiting even a few seconds feels annoying. The Extended Thinking version takes noticeably longer to respond because it deliberately works through problems step by step before giving an answer. This split approach mirrors a growing trend in the AI industry, where leading labs now offer separate modes for daily help and serious reasoning tasks.
Google says both models will reach consumers through the Gemini mobile app and developer application programming interfaces. The Extended Thinking mode is especially aimed at software coding, advanced math, scientific research, and complex business planning. By packaging the two options under the same version number, Google hopes to compete more directly with rivals that have already captured attention with specialized reasoning systems.
The move matters because transparent reasoning is becoming the new battleground for advanced AI. A model that can show its work and catch its own mistakes is far more useful for professionals than one that simply guesses the most likely next word. It also builds user trust, since people can follow how the AI reached a conclusion rather than receiving a black-box answer that might hide subtle errors.
Analysts expect Google to keep improving both tracks in parallel over the coming months. Future updates will likely shrink the speed gap while making the reasoning even sharper and more reliable. For now, the dual release sends a clear signal: Google wants to lead in both instant conversation and deep analysis, and it does not want enterprise customers or power users to leave its ecosystem for specialized rivals.