Gemini 4 Argon Is Here: What’s New in Google’s Latest AI Model and Who Can Use It
Quick answer: Gemini 4 Argon is Google’s newest AI model, available in the Gemini app from October 13, 2026 for Google AI Pro and Ultra subscribers. It brings stronger reasoning and coding, more direct answers, a new Tasks mode that keeps working when you step away, and a 1-million-token context window for developers.
Google has rolled out Gemini 4 Argon, the latest model in its Gemini family. Google describes it as bringing frontier-level reasoning and better performance on complex tasks — with a noticeable change in how it talks to you.
What’s new in Gemini 4 Argon
More direct answers
Google says Argon gives more direct responses with less hedging and repetition, and shares progress updates more often when working on longer tasks.
Big coding upgrade
Coding is one of the biggest improvements. In the Gemini app you can now switch between chatting and coding using Google Antigravity’s agentic coding interface, which has been added to the Gemini app for macOS. According to reports, Google has said Argon helped increase internal agentic code submissions by more than 35%.
Tasks: Gemini keeps working while you are away
A new Tasks toggle in the main chat window lets you hand off work to Gemini. It can continue working on a task even when you are not at your computer.
40+ connected apps
Gemini now connects with more than 40 apps, including Instacart, OpenTable, Adobe and Shopify, so it can take actions across services.
Who can use it?
At launch, Gemini 4 Argon and Tasks are available to Google AI Pro and Google AI Ultra subscribers in the Gemini app. Google’s announcement does not say whether free users get access, and it does not list country-by-country availability — Indian users can check the model picker in the Gemini app to see if Argon is offered on their plan.
For developers: Gemini 4 Argon in the API
| Detail | Gemini 4 Argon |
|---|---|
| Model ID | gemini-4-argon |
| Inputs | Text, images, video, audio, PDF |
| Output | Text |
| Context window | 1,048,576 tokens (about 1 million) |
| Default output limit | 65,536 tokens |
| Reported pricing | $2 per million input tokens, $10 per million output tokens |
Cached input tokens are reported to be about 95% cheaper. Google’s leadership has also said Argon ranks first on the Vals Index for coding, finance and legal work, according to reports. Check Google’s official pricing page before building, as prices can change.
Should you upgrade?
- Developers and coders: the agentic coding interface and long context window make it worth trying.
- Professionals: Tasks and connected apps can save time on multi-step work.
- Casual users: if you mainly ask quick questions, the free Gemini experience may be enough for now.
Frequently asked questions
Is Gemini 4 Argon free?
At launch it is available to Google AI Pro and Ultra subscribers. Google has not confirmed free-tier access.
What is the context window of Gemini 4 Argon?
About 1 million tokens (1,048,576), shared between input and output.
Does Gemini 4 Argon generate images?
The API model outputs text. It can understand images, video, audio and PDFs as input.
Based on Google’s Gemini app announcement (October 13, 2026), the Gemini API model page and industry reports.
