Gemini 3.6 Flash is pitched as cheaper and more token-efficient, but developers say coding has worsened

techartist_ · x · 2026-07-22

A reply thread around Gemini 3.6 Flash says the model is now positioned as more intelligent, more token-efficient, and cheaper, based on developer feedback.

A developer pushes back, saying coding feels worse right now on Gemini: too many bugs, and the Canvas preview does not even load for them.

Related event: Google Releases Three New Gemini Models, Focusing on Token Efficiency and Lower Cost(67 posts)→

Original post →

More from Models

Models channel →