Dev Tools · 1h ago
Gemini 3.6 Flash cuts token use 17%, lowers cost
Google's Gemini 3.6 Flash reduces output tokens by 17% versus 3.5 Flash, with pricing at $1.50/1M input and $7.50/1M output. The improvement is most notable on coding and web tasks, benefiting production agent systems. Vercel's AI Gateway now supports both new models, enabling easy migration and multi-provider routing.
Meridian48 take
The 17% token reduction compounds significantly in multi-step agent workflows, making this update more impactful than the headline suggests for teams already using Gemini.
Read the full reporting
Gemini 3.6 Flash: 17% fewer tokens, lower cost, and a Python cold start fix you didn't have to ask for →
DEV Community
gemini-3.6-flashai-gateway