Google begins rollout of Gemini 3.8 Live models

Google introduced Gemini 3.8 Live and an Extended Thinking variant for voice-based interaction and task execution.

By twenty8digital · AI-assisted briefingPublished 29 Sept 2026

Conceptual digital network illustration accompanying this technology brief

AI-generated conceptual illustration from the twenty8digital archive. Not a documentary image of this announcement.

Announcement: 15 Sept 2026
Coverage: Global
Primary source: Google ↗

What changed?

What should you keep in mind?

The models can continue a conversation while tools run in the background and can use visual context. Developer rollout began through the Gemini API and AI Studio. Enterprise availability includes private preview, while consumer access varies by product and subscription. Google says generated audio carries SynthID watermarking.

Go to the source

Source announcement: 15 Sept 2026. This is an original summary, not independent on-the-ground reporting.

Google ↗

Last updated: 29 Sept 2026. Editorial policy & corrections

Google begins rollout of Gemini 3.8 Live models

What changed?

What should you keep in mind?

Go to the source

Keep your finger on the pulse.

npm adds opt-in tag management to trusted publishing

GitHub Team gains self-serve Advanced Security trials

GitHub warns of October 7 TLS change for data-residency customers