Business

Gemini 4 Argon is Google’s AI built for the big jobs

By Sohail Khan 1 October 2026, 10:52 am

Benchmarks
Nifty22,556.85-63.61

FEATURED FUNDS★★★★★

Motilal Oswal Midcap Fund Direct-Growth

5Y Return
19.86 %

Invest Now

Enter search text:

Business News›AI›AI Insights›Google’s new Gemini 4 Argon targets complex coding, finance and legal work

Read Today's Paper

Google’s new Gemini 4 Argon targets complex coding, finance and legal work
Synopsis

Google has introduced Gemini 4 Argon, a new AI model aimed at improving complex reasoning and enterprise functionalities. This model is capable of processing up to 1 million tokens for greater computational tasks. Google is initially testing Argon with trusted cybersecurity defenders, working on significant projects for code migration.

FILE PHOTO: Google Gemini lettering, during the opening of Google's new Artificial Intelligence (AI) centre in Berlin, Germany, March 5, 2026.

Google has launched Gemini 4 Argon, its latest frontier artificial intelligence (AI) model, with a focus on complex reasoning, long-running tasks, software engineering, and enterprise work.

In its announcement, Google said that the model is designed to handle workflows that require AI systems to work through multiple steps rather than simply answer individual prompts. Google is initially rolling it out to a group of trusted cybersecurity defenders through its Fairwind Program, before making it available more broadly to developers, enterprises and consumers.

A much longer output window

One of the biggest changes in Argon is the amount of information the model can process and generate in a single trajectory.

Google has increased its output token limit to 1 million tokens, up from 64,000 tokens previously. This gives Argon considerably more room to reason through complex problems, write large amounts of code or work through lengthy research tasks without having to break them into multiple interactions.

Also Read: FTC opens probe into AI giants including Anthropic and OpenAI, source says

Google says this is particularly useful for long-horizon tasks where an AI agent may need to plan, execute, check its work, and continue working over many steps.

Coding and software engineering

Google says the model is already being used internally by thousands of employees for debugging, algorithm design, and large-scale code migrations. It is also working on migrating C and C++ codebases to the Rust programming language, including projects ranging from tens of thousands of lines of code to more than 800,000 lines in the Fuchsia Zircon kernel.

Beyond coding

Google is positioning Argon as an enterprise model rather than simply a programming assistant. The company says that it leads the Vals Index, which measures AI performance across areas including finance, coding, legal, and tax work.

The model also has strong multimodal capabilities. Google says Argon can analyse professional charts, understand long videos, and use information across multiple documents.

Built for cybersecurity

Google said the model can autonomously find, validate, and patch critical software vulnerabilities. The tech giant added that it is also working with cybersecurity company Wiz, which is using Argon through its Scan for Good initiative to identify and remediate exposures in critical infrastructure.

For trusted cybersecurity defenders and its own internal teams, Google says Argon will be released without cyber guardrails so they can use its full capabilities for defensive work.

Also Read: OpenAI launches Dots to take on Meta’s Muse with AI agents that can act on your behalf

What is Google doing about safety?

The increased autonomy of Argon also means Google is putting additional safeguards around the model before a wider rollout.

The company says it is strengthening protections against misuse, including cyber and chemical, biological, radiological, and nuclear risks. It has also worked on making the model more resistant to indirect prompt injection attacks, where malicious instructions embedded in external content can attempt to manipulate an AI agent.

Google is also deploying systems to monitor Argon's reasoning and actions for signs of misalignment — essentially, behaviour that goes beyond what the user intended. The system can stop execution when necessary.

The company says similar monitoring was used during training runs, with potential incidents sent to a dedicated response team. Google says it deliberately avoided feeding those findings back into training to reduce the risk of the model learning to evade the monitoring.

When will Gemini 4 Argon be available?

For now, Argon is being released in phases, starting with trusted cyber defenders and testers. Google says it is participating in the US government's voluntary pre-release model access process while using feedback from early users to improve its safeguards.

The model will eventually be available to developers, enterprises and consumers, starting with paid API customers and Google AI Ultra subscribers.

Google is pricing Argon at an introductory $2 per million input tokens and $10 per million output tokens, with cached input tokens priced at 95% below the standard input rate.

Add as a Reliable and Trusted News Source
Add Now!

Enter search text:

Continue reading with one of these options:
Limited Access
Free
Login to get access to some exclusive stories

& personalised newsletters
Login Now

Unlimited Access
Starting @ Rs120/month
Get access to exclusive stories, expert opinions &

in-depth stock reports
Subscribe Now

Unlock Access to All Exclusives with ETPrime FREE TRIAL

15 Days Free: Unlock All ETPrime Exclusives, Market Tools & ePapers

Trial offer expiring in00 : 05 : 00

I'd rather pay ₹2599 now

Uh-oh! This is an exclusive story available for selected readers only.

Worry not. You’re just a step away.

Prime Account Detected!

It seems like you're already an ETPrime member with

Log out of your current logged-in account and log in again using your ET Prime credentials to enjoy all member benefits.

Unlock ETPrime's Exclusive Stories Today!

Subscribe to gain powerful insights on business, stock market and industry trends.

Big Price Drop! Flat 40% Off

Student Only Offer

Already a Member? Sign In now

Already a Member? Sign In now
Already a Member? Sign In now

Offer Exclusively For You

Save up to Rs. 700/-

ON ET PRIME MEMBERSHIP

Avail Offer

Offer Exclusively For You

Get 1 Year Free

With 1 and 2-Year ET prime membership

Avail Offer

Offer Exclusively For You

Get 1 Year Free

With 1 and 2-Year ET prime membership

Avail Offer

Offer Exclusively For You

Get Flat 40% Off

Then ₹ 1749 for 1 year

Avail Offer

Offer Exclusively For You

ET Prime at ₹ 49 for 1 month

Then ₹ 1749 for 1 year

Avail Offer

Special Offer

Get flat 40% off on ETPrime

Avail Offer
Avail Offer

Claim NowAlready a Member? Sign In now

Need Assistance?

Connect with an ETPrime expert for personalized support and guidance

+91 9484700004

' + h3 + '

' + h4 + '

' + listing + '

What’s Included with

ETPrime Membership

1Exclusive Insights That Matter

Uncover the truth with our investigative stories

Make strategic moves using the real-world case studies

Read industry-specific stories to identify emerging trends

Spot opportunities with in-depth insights that matter

Trump temper on H-1B visas is forcing Indians to do these things to stay put in US

What Adani’s US indictment means for India Inc’s overseas fundraising
Why veterans like Reliance, L&T are on acquisition spree? Aswath Damodaran has an answer.

Will China’s dollar bond sale in Saudi Arabia trump the US in financial world?
Huawei launches its own OS to compete with Google and Apple. But can it win beyond China?

The problem with lab grown diamonds
Why a falling rupee is a better option for the economy

A list of top 20 momentum stocks that have delivered massive returns in one year