At a Glance

• Google has released Gemini 4 Argon globally for developers and businesses, targeting complex, multi-step tasks across coding, finance, law, cybersecurity and science.
• API pricing starts at $2 per million input tokens and $10 per million output tokens, with prices scheduled to double after November 30, 2026.
• Early users include Deloitte, LexisNexis, Ambience Healthcare and Isomorphic Labs.
• Google is expanding access to cybersecurity specialists through its Fairwind program and offering 10,000 scientists six months of free access to the highest usage tier of Antigravity.

---

Google, owned by Alphabet, has launched its flagship artificial intelligence model, Gemini 4 Argon, for developers and businesses worldwide, positioning it to compete more aggressively with OpenAI and Anthropic. The company says the model is designed for complex, multi-step work, including software development, finance, law, cybersecurity and scientific research.

---

Key Statements and Focus Areas

Advanced AI for Complex Tasks

Google describes Gemini 4 Argon as a model built for “deep work” involving complex, long-running tasks that require reasoning across multiple steps.

The company says it is designed to support real-world software engineering, enterprise knowledge work and cybersecurity defense. Google also claims the model performs strongly on industry benchmarks, including the Vals Index, which evaluates AI performance in areas such as finance, coding, legal work and taxation. These performance claims come from Google and should be distinguished from independently verified assessments.

Pricing and Availability

At launch, Gemini 4 Argon is available to developers and businesses through the Gemini API. Google AI Pro and Ultra subscribers can also access it through the Gemini app and AI Mode in Search, while access through Google Antigravity's agent-focused coding platform is limited to Ultra subscribers.

The introductory API price is $2 per million input tokens and $10 per million output tokens. From December 1, the listed rates are scheduled to rise to $4 per million input tokens and $20 per million output tokens.

Early Business Results

Google says teams using Argon through Antigravity increased code submissions by more than 35% while reducing code rollbacks by approximately one-third.

The company has named Deloitte, LexisNexis, Ambience Healthcare and Isomorphic Labs among organizations that tested the model during its early-access phase. The reported productivity figures are company-provided results, rather than a universal measure of performance across all customers.

Cybersecurity and Scientific Research

Google is expanding its Fairwind program to hundreds of additional vetted cybersecurity partners. Participants receive access to a version of Argon without some of its built-in cybersecurity restrictions, with the stated aim of helping specialists identify and address vulnerabilities.

Separately, Google says 10,000 scientists will receive six months of free access to the highest usage tier of Antigravity, supporting research that may benefit from advanced AI tools.

Google’s Competitive Position

The launch follows a difficult period for Google’s AI development efforts. In September, the company dropped plans to release Gemini 3.5 Pro, which CEO Sundar Pichai had initially indicated would arrive in June.

The release of Argon comes amid competition with OpenAI and Anthropic, as leading AI developers seek to improve model reasoning, coding performance and the ability to complete complex tasks with limited human intervention. Google has also made changes to its AI organization and expanded testing before the wider release of its latest model.

Google has released Gemini 4 Argon to developers and businesses worldwide, introducing a new flagship AI model designed to handle complex tasks across several professional fields.

The company says Argon is built for work requiring sustained reasoning and multiple steps, including software engineering, financial analysis, legal work, cybersecurity and scientific research. Google is making the model available through its API and selected paid subscription services.

The launch pricing starts at $2 per million input tokens and $10 per million output tokens. Those rates are scheduled to double after November 30, making the introductory pricing period a significant consideration for developers and businesses evaluating the model.

Google has also reported productivity gains among teams using Argon through Antigravity, its AI-focused development platform. According to the company, those teams increased code submissions by more than 35% and reduced code rollbacks by roughly one-third. Deloitte, LexisNexis, Ambience Healthcare and Isomorphic Labs are among the organizations that tested the model early.

Beyond commercial use, Google is expanding the Fairwind program to give more vetted cybersecurity professionals access to a version of Argon with certain cyber safeguards removed. The initiative is intended to help defenders examine systems and identify vulnerabilities, although access is restricted to approved participants.

The company is also offering 10,000 scientists six months of free access to Antigravity’s highest usage tier, extending the model’s potential applications into research.

The release comes after delays and organizational changes at Google’s DeepMind AI laboratory, as the company seeks to strengthen its position against OpenAI and Anthropic. While Google has promoted Argon’s performance on complex tasks, its wider impact will depend on independent evaluations, customer adoption, cost and reliability in real-world use.

FYI

Gemini 4 Argon’s launch reflects a shift in competition among leading AI developers: the focus is increasingly on systems that can carry out extended workflows, rather than simply generate text or answer isolated questions. Stronger performance in coding, legal analysis, finance and cybersecurity could make these models more useful to businesses, but it also raises questions about reliability, oversight, data security and the costs of sustained use.

For Google, Argon is also a strategic release following delays in its model roadmap and changes within DeepMind. Broadening access through paid subscriptions, developer APIs, cybersecurity partnerships and scientific research gives the company several routes to demonstrate the model’s value. However, productivity improvements reported by Google should be evaluated against independent testing and results from a wider range of customers.

The Fairwind program highlights the dual-use nature of advanced AI cybersecurity capabilities. Removing some restrictions for vetted defenders could help identify weaknesses before attackers exploit them, but such access also requires careful screening and safeguards. Ultimately, Argon’s significance will depend not only on benchmark performance, but on whether it can deliver measurable, dependable benefits while managing the risks associated with more capable AI systems.

Dive Deeper: Handpicked Stories for You

Follow Channel8 for continuous updates:

Three Silicon Valley Engineers Charged with Stealing Google Trade Secrets for Iran

Iraq Partners with Google to Offer Free AI Tools to University Students

Google Pays €326 Million to Settle Tax Dispute with Italian Authorities