Gemini 3.6 Flash, 3.5 Flash Lite, and reasoning toggles
Create an agent or open an existing agent in the builder, then follow the steps below. Model and feature availability may depend on your plan.
What it does
Two new Gemini models joined the selector, and reasoning became available on Gemini 3.5 Flash and 3.1 Pro Preview.
Before this
Two additions and one control. The models are additions to the selector; the reasoning toggles were not — Gemini 3.5 Flash and 3.1 Pro Preview were already selectable and had no reasoning dropdown, so an existing agent could be made to think harder without changing model.
Why it matters
Flash Lite is the cheapest way to run high-volume classification and routing; the reasoning toggle lets one model cover both quick replies and harder work.
What changed
Two model additions and one control that applied to models already in the list.
- Gemini 3.6 Flash
- 1,048,576-token context, 65,536-token output, text and images, Actions supported. Gold and Pro.
- Gemini 3.5 Flash Lite
- The same window and output at the cheapest tier in the Gemini family, on every plan — tagged as the cheapest model in the selector.
- Reasoning toggles
- Gemini 3.5 Flash and 3.1 Pro Preview gained the Reasoning dropdown, so an existing agent could be made to think harder without changing model.
Flash Lite is the one to test on classification, routing, and short-reply agents: it is the cheapest option that still takes a million tokens of context and still calls Actions.