Google Introduces Next-Gen 'Gemini 3.8 Live' and 'Extended Thinking' Models
Google has unveiled 'Gemini 3.8 Live,' optimized for scalability, speed, and cost-efficiency, along with the 'Extended Thinking' model. New features include rea
On September 15, 2026, Google announced the next-generation 'Gemini 3.8 Live' model, designed to significantly improve scalability, speed, and cost-efficiency. This latest model offers an enhanced conversational AI experience with improved real-time interaction capabilities.
Image source: Google AI
Key Features of Gemini 3.8 Live
Gemini 3.8 Live introduces several features to elevate user interaction:
- Mid-Sentence Interruption Support: The model supports 'mid-sentence interruptions,' allowing for natural conversation flow even when users interject.
- Real-Time Language Switching: Gemini 3.8 Live supports real-time switching across 97 different languages, facilitating seamless communication.
- Visual Context Awareness: The model leverages camera input to perceive and understand surrounding environments or screen content in real-time, enabling context-aware responses.
Additionally, the newly announced 'Extended Thinking' model provides enhanced reasoning and complex cognitive abilities, assisting users in solving intricate problems more effectively.
Practical Impact and Availability
Gemini 3.8 Live is engineered for maximum cost-efficiency, making it suitable for a wide range of applications. Detailed technical specifications and information regarding API availability and integration will be clarified in forthcoming technical documentation.