Google's Gemini 4 Argon Brings Advanced AI Reasoning to Developers and Subscribers

[jiaolix]

Google has released Gemini 4 Argon, its newest reasoning model, making it available globally on October 13 across the Gemini app, Google AI Studio, the Gemini API, and Google Cloud. The launch targets developers, businesses, and Pro and Ultra subscribers, with Google highlighting strong performance in coding, legal, financial, cybersecurity, and scientific workflows.

Where Gemini 4 Argon Is Available

According to Google's October 13 announcement, Gemini 4 Argon is offered to Pro and Ultra subscribers through the Gemini app and Search AI Mode. Developers can access the model through Google AI Studio and the Gemini API, while businesses can use it via Google Cloud. Google AI Ultra customers can also run Argon through Antigravity, the company's agent-based software development environment.

Google set introductory pricing at $2 per million input tokens** and **$10 per million output tokens. Per the company's developer documentation, both rates double after November 30, when the introductory period ends.

The model's specifications list a context window of 1,048,576 tokens and multiple levels of reasoning effort. While Argon's output is text, Google says it accepts text, images, video, audio, and PDF documents as inputs.

Coding, Reasoning, and Benchmark Results

In internal tests, Google reported over 35% more agentic code submissions through Antigravity. The company also said teams cut code rollbacks by almost a third when using Argon for engineering, and reported a 40% reduction in session preparation time for its Agent Development Kit.

Google stated that Argon took the top spot on the Vals Index, which measures how AI performs for professionals. Finance, legal work, coding, and tax work were among the strongest areas. A model card detailing safety testing, evaluation methods, and limitations accompanies the release.

The Gemini 4 Argon launch arrives five days after Anthropic released Claude Haiku 5.5, a smaller model built for high-volume tasks like customer support, priced at $0.10 per million input tokens for prompts under 100,000 tokens.

Cybersecurity and Scientific Research Use Cases

On the cybersecurity side, Google says Argon has helped researchers find and fix software defects in a safe environment. Through Google's Fairwind Program, Wiz examined vulnerabilities in healthcare and government systems, and the program expands to hundreds more vetted security partners beginning October 13.

In scientific research, Google announced that Argon sped up molecular modeling calculations and supported quantum computing experiments. Researchers at Isomorphic Labs used the model to review patents and scientific documents for drug discovery work.

Google said its safety assessments covered biological misuse, cyber threats, and unauthorized use of autonomous agents. In testing, the company reduced successful prompt-injection attacks to 0.7% across 15 adaptive attempts. Google also opened applications for 10,000 scientists to receive six months of free Antigravity access.

The launch follows Anthropic's $150 million commitment to federal AI research, announced on October 8 as part of its participation in the US Genesis Mission.

Getting Started With Gemini Subscriptions

With Gemini 4 Argon now rolling out to paid tiers, interest in Gemini plans has grown among users who want access to the latest reasoning features. Marketplaces such as z2u.com offer a straightforward way to secure a Gemini AI Subscription at competitive prices. Users comparing options can find a Buy Gemini Advanced plan or a Buy Gemini Subscription Key through z2u.com, making it easier to unlock Pro and Ultra capabilities without paying full retail rates.