Etavrian
keyboard_arrow_right Created with Sketch.
News
keyboard_arrow_right Created with Sketch.

Google Just Swapped Your AI to Gemini 3 Flash - Here's What Changes Overnight

Reviewed:
Andrii Daniv
2
min read
Dec 18, 2025
Minimalist AI control hub with toggle showing overnight switch to new AI and cost tradeoffs

Google has launched Gemini 3 Flash and is making it the default model in the Gemini app for users worldwide. The model is also rolling out globally as the standard system behind AI Mode in Google Search. Gemini 3 Flash is part of Google's Gemini 3 family, which also includes Pro and Deep Think variants.

Google Gemini 3 Flash becomes default in Gemini app & AI Mode
Gemini 3 Flash is now the default model in the Gemini app and in AI Mode in Google Search.

Gemini 3 Flash now default in Gemini app and AI Mode

According to Google's official announcement, Gemini 3 Flash replaces Gemini 2.5 Flash as the Gemini app's default model. The change applies globally, including for users on the free Gemini plan, who now receive Gemini 3 Flash responses by default without adjusting any settings.

  • Gemini 3 Flash is live in the Gemini app as the new default, replacing Gemini 2.5 Flash.
  • The rollout is global across supported regions where the Gemini app is available.
  • In Search, Gemini 3 Flash begins rolling out as the default model powering AI Mode responses.
  • Gemini 3 Pro is also available in the United States through the AI Mode model menu in Search.

Developer access and Gemini 3 Flash pricing

Gemini 3 Flash is available in preview for developers through the Gemini API. Access options include Google AI Studio, Google Antigravity, Vertex AI, and the Gemini Enterprise environment. Developers can also use the model through tools such as Gemini CLI and Android Studio integrations.

Google's Gemini API pricing documentation lists Gemini 3 Flash at $0.50 per million input tokens and $3.00 per million output tokens. For comparison, Gemini 2.5 Flash is listed at $0.30 per million input tokens and $2.50 per million output tokens.

Google reports that Gemini 3 Flash uses about 30% fewer tokens than Gemini 2.5 Pro for typical tasks. The company also cites external tests indicating that Gemini 3 Flash runs roughly three times faster than Gemini 2.5 Pro. These performance figures are drawn from Google's materials and referenced third-party measurements.

Background on the Gemini 3 model family

Gemini 3 Flash extends Google's Gemini 3 model family, which Google recently introduced for Search and related products. That earlier rollout previewed Gemini 3 Pro and announced Gemini 3 Deep Think as a reasoning-focused mode. Gemini 3 Flash now joins Pro and Deep Think as part of the broader Gemini 3 lineup and becomes the standard model behind the Gemini app and AI Mode in Search.

Official sources

This report is based on Google's official product materials for Gemini 3 Flash and other Gemini 3 announcements, including the product blog post in which Google released Gemini 3 Flash.

Quickly summarize and get insighs with: 
Author
Etavrian AI
Etavrian AI is developed by Andrii Daniv to produce and optimize content for etavrian.com website.
Reviewed
Andrew Daniv, Andrii Daniv
Andrii Daniv
Andrii Daniv is the founder and owner of Etavrian, a performance-driven agency specializing in PPC and SEO services for B2B and e‑commerce businesses.
Quickly summarize and get insighs with: 
Table of contents