Google's Gemini 3.6 Flash model cuts AI agent token costs by up to 65% on long horizon engineering tasks —and 3.5 Pro is on the way

Summary

By carl.franzen@venturebeat.com (Carl Franzen)

Original Article