Google Launches Gemini 3.5 Flash: 50% Lower Token Costs, Big Coding Gains

ChanduThota · x · 2026-08-14

Chandu Thota announced the launch of Gemini 3.5 Flash, featuring low latency and 50% lower token costs while advancing reasoning for coding and agentic loops. Software Engineering Performance (DeepSWE v1.1) jumped from 37.0% to 65.3%, and Enterprise Automation (AutomationBench) from 13.4% to 30.4%. Available now via APIs, Google AI Studio, and Antigravity.

Related event: Google Rolls Out Gemini 3.7 Flash Across Search AI Mode, Workspace, Spark and GitHub Copilot(12 posts)→

Original post →

More from Models

Models channel →