ToolsJuly 21, 2026via Vercel Blog
Gemini 3.6 Flash and Gemini 3.5 Flash-Lite are now available on AI Gateway
Why it matters
Google's latest Gemini models are shipping through Vercel's unified API layer, giving developers a zero-markup path to production AI. This signals how model access is shifting from direct provider APIs to infrastructure-as-middleware plays.
Key signals
- Gemini 3.6 Flash available on AI Gateway with improved coding and agentic task performance
- Gemini 3.5 Flash-Lite upgraded for subagent use cases
- AI Gateway provides unified API with usage tracking, cost monitoring, retries, failover, and performance optimization
- Zero platform fee on inference including Bring Your Own Key (BYOK) requests
- Pricing reflects provider rates with no markup
- Gemini 3.6 Flash and Gemini 3.5 Flash-Lite now available on Vercel AI Gateway
- Gemini 3.6 Flash: improved coding, agentic tasks, web development; fewer tokens and model calls
- Gemini 3.5 Flash-Lite: upgraded agentic capabilities for subagents
- AI Gateway pricing: no platform markup, no platform fee on inference, BYOK support included
- AI Gateway features: custom reporting, Zero Data Retention, budgets for API keys, routing rules, performance optimizations
- Published July 20, 2026
The hook
Gemini 3.6 Flash is now live on Vercel's AI Gateway—cleaner code output, fewer token burns, built for agents.
and are now available on AI Gateway.Gemini 3.6 FlashGemini 3.5 Flash-Lite
Gemini 3.6 Flash improves quality across coding, agentic tasks, and web development while consuming fewer tokens and making fewer model calls. It produces cleaner web and app development output.
Gemini 3.5 Flash Lite upgrades…