Gemini 3.6 Flash goes live in Antigravity with 17% fewer output tokens

rseroter · x · 2026-07-22

Gemini 3.6 Flash is now live in Antigravity.

According to the post, the new version builds on feedback from 3.5 Flash and uses up to 17% fewer output tokens while completing complex workflows with fewer reasoning steps and tool calls. The attached image also shows the Gemini 3.6 Flash branding.

This is a model-level update with a clear efficiency angle, not just a UI announcement.

Related event: Google Launches Three New Gemini Models with Higher Efficiency(65 posts)→

Original post →

More from Models

Models channel →