0% read

Gemini 4 Argon in the Gemini App: Access Guide

Oct 5, 2026

Google has not confirmed Gemini 4 Argon in the Gemini app model picker as of the October 5, 2026 access snapshot. Google’s Argon announcement says wider access starts with paid API customers and Google AI Ultra subscribers, but it does not publish an app release date or promise a specific model label in the consumer interface.

What Google’s October 9 plan update says

Google’s Gemini Apps Help update says model access changes begin October 9 for users without a paid AI plan. Its published plan table lists:

PlanModels listed after the change
No paid AI planFlash-Lite
AI PlusFlash-Lite and Flash
AI ProFlash-Lite, Flash, and Pro
AI UltraFlash-Lite, Flash, and Pro

Google says AI Plus changes follow timing sent by email. The labels Flash and Pro describe the app’s plan table; they do not prove which specific underlying model version powers each label, and the help page does not announce Gemini 4 Argon by name.

Why a paid plan is not an Argon confirmation

The app plan table and the Argon API rollout answer different questions. The Argon release announcement names paid API customers and AI Ultra subscribers for the wider rollout, with no date and no statement that every subscriber receives access at once. A subscription therefore does not establish that Argon is available in the website picker on a particular day.

For current access status, see who Google has confirmed for Argon. For the API rates that apply once access is available, see Argon pricing and limits.

App limits are not an Argon quota

The same Google help page says compute-based limits reset every five hours until a weekly limit, and that limits vary with complexity, features, and chat length. It does not publish an Argon-specific turn allowance. A question about the Gemini app’s plan limits should therefore not be turned into a claim about Argon’s quota.

What about Gemini 3.6 Flash and 3.7 Flash?

As checked on October 5, Google’s Gemini API deprecations page listed no announced shutdown date for gemini-3.6-flash or gemini-3.7-flash. App plan changes and API endpoint deprecations are separate systems. An app model label changing does not establish that either API endpoint is shutting down.

Product notice titled Gemini 3.6 and 3.7 Flash Deprecation, telling users to switch to Gemini 3.8 Flash.

Source: @HarshithLucky3. The notice in that screenshot says to use Gemini 3.8 Flash and that Gemini 3.6 Flash and Gemini 3.7 Flash will be turned down shortly. It does not name Argon, it does not show the Gemini app model picker, and it is not the API deprecations page linked above. It also does not give a calendar date.

For model performance, use the source-separated Argon benchmark table. For a model you can download and run locally, see Gemma 4; it is a different model family.

gemma4 — interact

Try Gemma 4 online

~/gemma4 $ Try Gemma 4 in the browser playground before setting up a local installation.

Open Gemma 4 playground />
Gemma 4 AI

Gemma 4 AI

Related Guides