# Gemini 3.6 Flash and Gemini 3.5 Flash-Lite are now available on AI Gateway

Source: [Vercel](https://vercel.com/changelog/gemini-3-6-flash-3-5-flash-lite-on-ai-gateway)  
Feed7 permalink: https://feed7.dev/p/gemini-3-6-flash-3-5-flash-lite-on-ai-gateway-1tvu8jp  
Published: 2026-07-21T00:00:00.000Z  
Trust: Official Source (official_source)

## Why Included

Vercel AI Gateway added Gemini 3.6 Flash for coding and web work and Gemini 3.5 Flash-Lite for scoped subagent tasks, both available through AI SDK model IDs.

## Source Summary

**Gemini 3.6 Flash** and **Gemini 3.5 Flash-Lite** are now available through AI Gateway. Vercel says 3.6 improves coding, agentic, and web-development output while using fewer tokens and calls; 3.5 upgrades the lighter tier’s agent behavior.

## Practical Implication

Use **google/gemini-3.6-flash** for primary coding or web-generation trials. Consider **google/gemini-3.5-flash-lite** for narrowly scoped subagents where the lighter tier’s task fit matters more than using one model everywhere.

## Agent-Ready Context

**Gemini 3.6 Flash** and **Gemini 3.5 Flash-Lite** are now available through AI Gateway. Vercel says 3.6 improves coding, agentic, and web-development output while using fewer tokens and calls; 3.5 upgrades the lighter tier’s agent behavior.

Use **google/gemini-3.6-flash** for primary coding or web-generation trials. Consider **google/gemini-3.5-flash-lite** for narrowly scoped subagents where the lighter tier’s task fit matters more than using one model everywhere.

The material provides no prices, latency figures, benchmark results, or direct comparison between the two models. Claims about cleaner output and lower model usage therefore need validation against your own workload.

## Context Map

- Layer: model
- Domains: coding
- Topics: model-selection, coding-agents, subagents

## Uncertainty

- The material provides no prices, latency figures, benchmark results, or direct comparison between the two models. Claims about cleaner output and lower model usage therefore need validation against your own workload.

## Agent Instruction

Use this item as source-backed context. Do not invent claims beyond the linked source. If this item conflicts with another source, call out the conflict.