Google bets on Gemini 4 Argon to regain lost ground in frontier AI race- Moneycontrol.com

October 14, 2026 / 00:22 IST

Market Mastery

Webinar by Vishal Malkan

Find the weak links

in your portfolio by Vishal Malkan

Google seeks to regain ground in the fiercely competitive frontier AI race with the release of Gemini 4 ‘Argon’, the latest version of its flagship AI model, after trailing rivals for months and a major shakeup in its AI unit.

The model, first introduced a fortnight ago to a set of trusted cybersecurity experts, is now available globally to Gemini app users, developers, and enterprises.

The company claims that Argon outperforms frontier models from rivals including OpenAI's GPT-6 Astra, and Anthropic's Claude Fable 5.1, and Claude Opus 5.5 across a series of benchmarks.

The model "excels at deep work" in complex, real-world tasks across domains including coding, finance, legal, cybersecurity, and science, it said.

Argon will be available at an introductory price of $2 per million input tokens and $10 per million output tokens, making it one-fifth the cost of OpenAI’s GPT-6 Astra and Anthropic’s Fable 5 models.

The pricing is also half that of Meta’s Muse Wonder AI model, released yesterday. However, rates are set to double after November 30, that will bring Argon’s pricing on a par with Meta’s model.

Google DeepMind overhaul

The release comes after a major overhaul in Google DeepMind with Demis Hassabis stepping down as CEO to become its Chair and Chief Scientist at Alphabet, Google's parent company.

Jeff Dean, chief scientist of Google DeepMind and Google Research, also left after a 27-year stint to launch Discovery Loop, a public benefit corporation, alongside fellow AI researchers Sanjay Ghemawat, Oriol Vinyals, and Quoc Le.

The tech giant has also lost several senior AI researchers in recent months, including Gemini co-lead Noam Shazeer, who joined OpenAI, and Nobel laureate John Jumper, who left for Anthropic.

While Google released several versions of its cost-efficient Flash models in quick succession, the company faced delays in launching its frontier model Gemini 3.5 Pro, which was announced in May for a June release before the launch was eventually scrapped.

The setback raised concerns that Google could lose ground to OpenAI and Anthropic, both of which continue to roll out new frontier AI models at a rapid pace.

How Gemini 4 Argon is being used?

Koray Kavukcuoglu, SVP, Google DeepMind said that Google's internal teams as well as enterprises and research institutions have been using Argon to identify and patch critical cybersecurity vulnerabilities; tackle complex engineering challenges; and optimise business workflows in areas such as legal, finance and science.

Argon in Antigravity has helped teams increase agentic code submissions by more than 35 percent while reducing code rollbacks by nearly a third, he said. It has also made Google's Gemma models nearly three times as fast, he added.

Kavukcuoglu also noted that the model has been used to advance research in quantum computing, nuclear fusion and life sciences.

To aid this pace of research, Google will offer 10,000 scientists six months of free access to its highest usage tier in Antigravity, the company said.

Google is also expanding its Fairwind programme, that gives cybersecurity defenders early access to the firm's advanced models, to hundreds of additional partners.

The model is currently rolling out to Google Ultra and Pro subscribers in the Gemini app and Search and Ultra subscribers in Google Antigravity. For developers and enterprises, it’s available in the paid tier of the Gemini API and the Gemini platform.