<p>Google has unveiled Gemini 4 Argon, its latest frontier AI model, but most users cannot access it yet. The model is initially being rolled out to a small group of trusted cyber defenders through Google's Fairwind Programme, as the company tests its safety systems before a wider launch.</p><p>&nbsp;</p><p>The restricted release makes Argon unusual even in the fast-moving AI market. Google says the model is capable of handling complex, long-running tasks across software engineering, legal and financial work and cybersecurity.&nbsp; It also comes with a major technical upgrade: a 1 million-token maximum output limit, up from 64,000 tokens on previous Gemini models. That gives Argon considerably more room to reason through lengthy, multi-step tasks in a single run.</p><p>&nbsp;</p><h2><strong>Why Google is keeping Gemini 4 Argon restricted</strong></h2><p>Google says the phased launch is necessary because frontier AI capabilities can create new safety and security risks. The company is participating in a US government voluntary process for pre-release access and plans to use feedback from early testers to improve its guardrails. Argon has been trained for cybersecurity defence and can autonomously identify, validate and patch software vulnerabilities, according to Google. The model is already being used internally. Google says Argon agents analysed data-centre telemetry and identified memory optimisations that could free more than 300 TiB of memory once deployed, with estimated total savings of 500 TiB to 1 PiB.</p><p>&nbsp;</p><h2><strong>Gemini 4 Argon vs OpenAI and Anthropic</strong></h2><p>Google's published benchmarks show a mixed picture rather than a clean sweep. Argon scores 68.9 per cent on the Vals Index, compared with 67.0 per cent for Claude Opus 5.5. It also scores 77.9 per cent on DeepSWE v1.1, ahead of Opus 5.5's 74.2 per cent. But rivals still lead on some tasks. Claude Opus 5.5 scores 66.4 per cent on Terminal-Bench 4.0, compared with Argon's 57.4 per cent. GPT-6 Astra also leads Argon on several other evaluations, including FrontierSWE v2 and OSWorld-2.0. That means Google's results point to a highly capable model, but not one that dominates every workload.</p><p>&nbsp;</p><h2><strong>Gemini 4 Argon price and wider release</strong></h2><p>Google says Argon will initially cost $2 per million input tokens and $10 per million output tokens. Cached input is priced 95 per cent below the standard input rate. After the introductory period, prices rise to $4 and $20 respectively. For now, access remains limited to trusted cyber defenders. Google says broader availability will begin with paid API customers and Google AI Ultra subscribers before expanding further, although it has not given a specific public release date. The bigger question is therefore not simply whether Gemini 4 Argon is powerful. It is whether Google's combination of long-context reasoning, agentic capabilities and competitive pricing can turn that capability into a meaningful advantage once the model reaches more developers and businesses.</p>

Google's Gemini 4 Argon is a new frontier AI model with a 1 million-token output limit and capabilities across coding, enterprise work and cybersecurity. It is initially restricted to trusted cyber defenders, with wider access planned for developers and consumers later.

Google has unveiled Gemini 4 Argon, its latest frontier AI model, but most users cannot access it yet. The model is initially being rolled out to a small group of trusted cyber defenders through Google's Fairwind Programme, as the company tests its safety systems before a wider launch. The restricted release makes Argon unusual even in the fast-moving AI market. Google says the model is capable of handling complex, long-running tasks across software engineering, legal and financial work and cybersecurity. It also comes with a major technical upgrade: a 1 million-token maximum output limit, up from 64,000 tokens on previous Gemini models. That gives Argon considerably more room to reason through lengthy, multi-step tasks in a single run. Why Google is keeping Gemini 4 Argon restricted Google says the phased launch is necessary because frontier AI capabilities can create new safety and security risks. The company is participating in a US government voluntary process for pre-release access and plans to use feedback from early testers to improve its guardrails. Argon has been trained for cybersecurity defence and can autonomously identify, validate and patch software vulnerabilities, according to Google. The model is already being used internally. Google says Argon agents analysed data-centre telemetry and identified memory optimisations that could free more than 300 TiB of memory once deployed, with estimated total savings of 500 TiB to 1 PiB. Gemini 4 Argon vs OpenAI and Anthropic Google's published benchmarks show a mixed picture rather than a clean sweep. Argon scores 68.9 per cent on the Vals Index, compared with 67.0 per cent for Claude Opus 5.5. It also scores 77.9 per cent on DeepSWE v1.1, ahead of Opus 5.5's 74.2 per cent. But rivals still lead on some tasks. Claude Opus 5.5 scores 66.4 per cent on Terminal-Bench 4.0, compared with Argon's 57.4 per cent. GPT-6 Astra also leads Argon on several other evaluations, including FrontierSWE v2 and OSWorld-2.0. That means Google's results point to a highly capable model, but not one that dominates every workload. Gemini 4 Argon price and wider release Google says Argon will initially cost $2 per million input tokens and $10 per million output tokens. Cached input is priced 95 per cent below the standard input rate. After the introductory period, prices rise to $4 and $20 respectively. For now, access remains limited to trusted cyber defenders. Google says broader availability will begin with paid API customers and Google AI Ultra subscribers before expanding further, although it has not given a specific public release date. The bigger question is therefore not simply whether Gemini 4 Argon is powerful. It is whether Google's combination of long-context reasoning, agentic capabilities and competitive pricing can turn that capability into a meaningful advantage once the model reaches more developers and businesses.

<p>Google's Gemini 4 Argon is a new frontier AI model with a 1 million-token output limit and capabilities across coding, enterprise work and cybersecurity. It is initially restricted to trusted cyber defenders, with wider access planned for developers and consumers later.</p>

Google Gemini 4 Argon is here, but few can use its most powerful AI yet

<p>Google has unveiled Gemini 4 Argon, its latest frontier artificial intelligence model, but most people cannot use it yet.</p><p>&nbsp;</p><p>The model is initially rolling out to a group of trusted cyber defenders through Google's Fairwind Program. Google says the restricted launch is part of a phased approach to testing the model's capabilities and strengthening safeguards before wider access. Argon is designed for long, complex tasks across software engineering, finance, legal work and cybersecurity. Google says it is already being used internally, where teams have used Argon agents to analyse data-centre memory and identify optimisations that could free up more than 300 TiB of memory once deployed.</p><h2><strong>Gemini 4 Argon gets a huge 1-million-token output limit</strong></h2><p>One of the biggest changes is the model's ability to generate up to 1 million tokens in a single output, compared with 64,000 previously. That matters because Argon is designed to work through lengthy, multi-step problems rather than simply answer individual prompts. Google says its engineers are using it for debugging, algorithm design and large-scale codebase migrations. The company also says Argon achieved a 77.9% score on DeepSWE v1.1, a benchmark for long-horizon software engineering tasks. Google reported leading results in several other professional and business evaluations as well.</p><p>&nbsp;</p><p>The model has also been used for large software migrations, including work moving C/C++ code towards Rust. Google says one Argon-assisted project produced a Rust video decoder that ran 2.7 times faster than an existing Rust implementation while producing identical video output.</p><h2><strong>Its cybersecurity ability is the reason for the caution</strong></h2><p>Argon's most sensitive capability may be cybersecurity. Google says the model can find, validate and patch critical software vulnerabilities autonomously. It is already being used by cybersecurity company Wiz through its Scan for Good initiative. That ability has an obvious dual-use problem. The same technology that can help defenders discover vulnerabilities could potentially be misused to find weaknesses for malicious purposes.</p><p>Google therefore plans to provide Argon without some cyber guardrails only to trusted defenders and internal teams. The company says it is continuing work on safeguards against misuse, prompt injection and model misalignment before expanding availability.</p><h2><strong>When will Gemini 4 Argon be available and how much will it cost?</strong></h2><p>Google says Argon will eventually become available to developers, enterprises and consumers, starting with paid API customers and Google AI Ultra subscribers. The introductory API price is $2 per million input tokens and $10 per million output tokens. Cached input tokens receive a 95% discount during the introductory period. After that period, Google says pricing will rise to $4 per million input tokens and $20 per million output tokens. For now, however, Gemini 4 Argon's most powerful capabilities remain behind a controlled rollout. Google says feedback from its first group of cyber defenders and trusted testers will help determine the next stage of availability.</p>

Google has launched Gemini 4 Argon, a frontier AI model designed for complex coding, enterprise work and cybersecurity. It supports up to 1 million output tokens and can autonomously find and patch vulnerabilities, but access is initially limited to trusted cyber defenders.

Google has unveiled Gemini 4 Argon, its latest frontier artificial intelligence model, but most people cannot use it yet. The model is initially rolling out to a group of trusted cyber defenders through Google's Fairwind Program. Google says the restricted launch is part of a phased approach to testing the model's capabilities and strengthening safeguards before wider access. Argon is designed for long, complex tasks across software engineering, finance, legal work and cybersecurity. Google says it is already being used internally, where teams have used Argon agents to analyse data-centre memory and identify optimisations that could free up more than 300 TiB of memory once deployed. Gemini 4 Argon gets a huge 1-million-token output limit One of the biggest changes is the model's ability to generate up to 1 million tokens in a single output, compared with 64,000 previously. That matters because Argon is designed to work through lengthy, multi-step problems rather than simply answer individual prompts. Google says its engineers are using it for debugging, algorithm design and large-scale codebase migrations. The company also says Argon achieved a 77.9% score on DeepSWE v1.1, a benchmark for long-horizon software engineering tasks. Google reported leading results in several other professional and business evaluations as well. The model has also been used for large software migrations, including work moving C/C++ code towards Rust. Google says one Argon-assisted project produced a Rust video decoder that ran 2.7 times faster than an existing Rust implementation while producing identical video output. Its cybersecurity ability is the reason for the caution Argon's most sensitive capability may be cybersecurity. Google says the model can find, validate and patch critical software vulnerabilities autonomously. It is already being used by cybersecurity company Wiz through its Scan for Good initiative. That ability has an obvious dual-use problem. The same technology that can help defenders discover vulnerabilities could potentially be misused to find weaknesses for malicious purposes. Google therefore plans to provide Argon without some cyber guardrails only to trusted defenders and internal teams. The company says it is continuing work on safeguards against misuse, prompt injection and model misalignment before expanding availability. When will Gemini 4 Argon be available and how much will it cost? Google says Argon will eventually become available to developers, enterprises and consumers, starting with paid API customers and Google AI Ultra subscribers. The introductory API price is $2 per million input tokens and $10 per million output tokens. Cached input tokens receive a 95% discount during the introductory period. After that period, Google says pricing will rise to $4 per million input tokens and $20 per million output tokens. For now, however, Gemini 4 Argon's most powerful capabilities remain behind a controlled rollout. Google says feedback from its first group of cyber defenders and trusted testers will help determine the next stage of availability.

<p>Google has launched Gemini 4 Argon, a frontier AI model designed for complex coding, enterprise work and cybersecurity. It supports up to 1 million output tokens and can autonomously find and patch vulnerabilities, but access is initially limited to trusted cyber defenders.</p>

Gemini 4 Argon: Google’s new AI can find and fix software flaws on its own

gemini-4-argon

Gemini 4 Argon, News on Gemini 4 Argon, stories on Gemini 4 Argon, articles on Gemini 4 Argon, Gemini 4 Argon stories at Wion, Gemini 4 Argon news, Gemini 4 Argon latest news

gemini 4-argon

Google Gemini 4 Argon is here, but few can use its most powerful AI yet

Gemini 4 Argon: Google’s new AI can find and fix software flaws on its own