Google Antigravity Blog: Gemini 3.6 Flash in Google Antigravity

The new model cuts output tokens by 17% versus 3.5 Flash, costs less per token, and improves coding and multimodal performance.

Costs less per token than 3.5 Flash
Takes fewer reasoning steps and tool calls for multi-step workflows
Delivers higher coding precision with fewer unwanted edits and reduced execution loops
Available through Google Antigravity 2.0

Developers using Google Antigravity get a cheaper, more token-efficient model that improves coding and agentic workflows.

Sources