--- title: Gemini 4 Argon: Google’s New AI Is Built to Think and... description: Google has introduced Gemini 4 Argon, a frontier AI model designed for long-horizon professional workflows rather than short chatbot-style exchanges. It raises... canonical: https://daily.dev/posts/gemini-4-argon-google-s-new-ai-is-built-to-think-and-work-for-much-longer-ibbgfhpdj twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Gemini 4 Argon: Google’s New AI Is Built to Think and Work for Much Longer | daily.dev og:description: Google has introduced Gemini 4 Argon, a frontier AI model designed for long-horizon professional workflows rather than short chatbot-style exchanges. It raises... og:url: https://daily.dev/posts/gemini-4-argon-google-s-new-ai-is-built-to-think-and-work-for-much-longer-ibbgfhpdj og:image: https://api.daily.dev/og/posts/iBbGfHPdj.png og:image:alt: Gemini 4 Argon: Google’s New AI Is Built to Think and Work for Much Longer og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Gemini 4 Argon: Google’s New AI Is Built to Think and Work for Much Longer **[Medium](https://daily.dev/sources/medium_js)** · 13 min read · 0 upvotes · 0 comments ## Summary Google has introduced Gemini 4 Argon, a frontier AI model designed for long-horizon professional workflows rather than short chatbot-style exchanges. It raises the output token limit from 64,000 to 1 million tokens, enabling extended reasoning, tool use, and iterative experimentation. Google cites internal use cases including large-scale C++-to-Rust migrations (including an 800,000-line Fuchsia Zircon kernel migration), a 2.7x speedup in its libgav1 video decoder, and over 300 TiB of data center memory freed through AI-driven optimization. The model is being piloted for autonomous defensive cybersecurity work through Google's Fairwind Program, with partner Wiz reportedly finding a critical healthcare vulnerability missed by prior models. The announcement also addresses prompt injection resilience, misalignment monitoring safeguards, enterprise benchmark results (AutomationBench, LVBench, DeepSWE v1.1), and new pricing: $2/$10 per million input/output tokens initially, rising to $4/$20 after an introductory period. Reuters notes Argon trails some competitors on certain coding benchmarks despite leading on others. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is the output token limit for Google's Gemini 4 Argon model? Gemini 4 Argon supports up to 1 million output tokens, a major increase from the previous frontier limit of 64,000 tokens. Google states this extra room gives the model more computational headroom to reason through difficult problems, generate intermediate work, and sustain long-horizon tasks like software engineering, research, and multi-step enterprise automation rather than just producing longer text. _daily.dev surfaces updates like this for engineers tracking how much agentic AI can realistically handle in production workflows._ ### How much does Google charge to use Gemini 4 Argon via API? Gemini 4 Argon is initially priced at $2 per million input tokens and $10 per million output tokens, with a discount for cached input. After the introductory period, pricing rises to $4 per million input tokens and $20 per million output tokens. This pricing structure matters because long-horizon tasks can generate hundreds of thousands of tokens, making inference cost a real factor in adopting agentic workflows. _Teams budgeting for agentic AI workloads can follow pricing shifts like this one on daily.dev._ ### What cybersecurity capabilities does Google's Gemini 4 Argon model have? Argon was trained specifically for defensive cybersecurity and can autonomously find, validate, and patch critical vulnerabilities. Through the Fairwind Program, Google is giving selected trusted defenders access without normal cyber guardrails. Cybersecurity firm Wiz reportedly used Argon via its Scan for Good initiative to discover a critical healthcare software vulnerability that previous frontier models had missed. _daily.dev helps security-minded developers keep up with autonomous vulnerability discovery tools like this one._ --- Tags: [#google](https://daily.dev/tags/google), [#llm](https://daily.dev/tags/llm), [#ai-agents](https://daily.dev/tags/ai-agents), [#google-gemini](https://daily.dev/tags/google-gemini), [#ai-security](https://daily.dev/tags/ai-security) [View this post on daily.dev](https://daily.dev/posts/gemini-4-argon-google-s-new-ai-is-built-to-think-and-work-for-much-longer-ibbgfhpdj) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Gemini 4 Argon: Google’s New AI Is Built to Think and Work for Much Longer","url":"https://daily.dev/posts/gemini-4-argon-google-s-new-ai-is-built-to-think-and-work-for-much-longer-ibbgfhpdj","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/gemini-4-argon-google-s-new-ai-is-built-to-think-and-work-for-much-longer-ibbgfhpdj"},"datePublished":"2026-10-04T03:57:47.074Z","dateModified":"2026-10-04T03:59:39.161Z","description":"Google has introduced Gemini 4 Argon, a frontier AI model designed for long-horizon professional workflows rather than short chatbot-style exchanges. It raises...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/d6ab0a9a223ee01bc82d647f54feb2d2?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/d6ab0a9a223ee01bc82d647f54feb2d2?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Medium","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Medium","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/medium","url":"https://daily.dev/sources/medium_js"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/gemini-4-argon-google-s-new-ai-is-built-to-think-and-work-for-much-longer-ibbgfhpdj","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"google,llm,ai-agents,google-gemini,ai-security","timeRequired":"PT13M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Medium","item":"https://daily.dev/sources/medium_js"},{"@type":"ListItem","position":3,"name":"Gemini 4 Argon: Google’s New AI Is Built to Think and Work for Much Longer"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/gemini-4-argon-google-s-new-ai-is-built-to-think-and-work-for-much-longer-ibbgfhpdj#faq","mainEntity":[{"@type":"Question","name":"What is the output token limit for Google's Gemini 4 Argon model?","acceptedAnswer":{"@type":"Answer","text":"Gemini 4 Argon supports up to 1 million output tokens, a major increase from the previous frontier limit of 64,000 tokens. Google states this extra room gives the model more computational headroom to reason through difficult problems, generate intermediate work, and sustain long-horizon tasks like software engineering, research, and multi-step enterprise automation rather than just producing longer text. daily.dev surfaces updates like this for engineers tracking how much agentic AI can realistically handle in production workflows."}},{"@type":"Question","name":"How much does Google charge to use Gemini 4 Argon via API?","acceptedAnswer":{"@type":"Answer","text":"Gemini 4 Argon is initially priced at $2 per million input tokens and $10 per million output tokens, with a discount for cached input. After the introductory period, pricing rises to $4 per million input tokens and $20 per million output tokens. This pricing structure matters because long-horizon tasks can generate hundreds of thousands of tokens, making inference cost a real factor in adopting agentic workflows. Teams budgeting for agentic AI workloads can follow pricing shifts like this one on daily.dev."}},{"@type":"Question","name":"What cybersecurity capabilities does Google's Gemini 4 Argon model have?","acceptedAnswer":{"@type":"Answer","text":"Argon was trained specifically for defensive cybersecurity and can autonomously find, validate, and patch critical vulnerabilities. Through the Fairwind Program, Google is giving selected trusted defenders access without normal cyber guardrails. Cybersecurity firm Wiz reportedly used Argon via its Scan for Good initiative to discover a critical healthcare software vulnerability that previous frontier models had missed. daily.dev helps security-minded developers keep up with autonomous vulnerability discovery tools like this one."}}]} ```