Google has introduced Gemini 3.8 Flash and Gemini 3.8 Flash Cyber, two new AI models designed for advanced reasoning, coding, agentic workflows and cybersecurity. Announced on September 2, 2026, Gemini 3.8 Flash is positioned as Google’s most intelligent Flash model yet. The company says it delivers stronger performance while maintaining the speed and introductory pricing of Gemini 3.7 Flash. Gemini 3.8 Flash Cyber is designed specifically for cybersecurity defenders and focuses on vulnerability discovery and automated patching. The standard model is available to developers, enterprises and eligible consumers, while Cyber access is restricted through Google’s Fairwind Program.What Makes Gemini 3.8 Flash Different?Gemini 3.8 Flash is built for tasks that require extended reasoning, software engineering and autonomous agents. Google says the model can perform additional reasoning steps and use tools repeatedly on complex tasks, potentially improving results but also increasing token usage. It supports text, images, video, audio and PDFs, with a context window of up to one million input tokens and 65,536 output tokens.Designed for long-horizon software engineering and codingSupports autonomous and agentic workflowsOffers low, medium and high thinking effort levelsSupports multimodal inputs including text, images, video and audioAvailable through Google AI Studio, Gemini API and other Google platformsGemini 3.8 Flash Cyber FocusesThe second model, Gemini 3.8 Flash Cyber, is focused on defensive cybersecurity applications. Google says it demonstrates frontier-level performance in vulnerability discovery and automated patching. The company reports that the model achieved a success rate above 70% on an internal vulnerability-discovery evaluation covering complex codebases across 20 programming languages.Google also reports that Gemini 3.8 Flash Cyber achieved a 47.2% pass@1 result on the external CWE-Bench patching benchmark, compared with 47.8% for a leading frontier model cited by Google. Access is being provided to trusted government authorities, critical infrastructure operators and software maintainers through the Fairwind Program.Also Read: Instagram AI Profiles Face New Reach Limits Without Proper DisclosurePricing And Availability DetailsGemini 3.8 Flash is available at an introductory API price of <head>.75 per million input tokens and .75 per million output tokens. Google says this pricing remains in place through December 31, 2026, after which the listed prices will increase to .50 and .50 per million tokens respectively.For consumers, Gemini 3.8 Flash is available to Google AI Pro and Ultra subscribers through supported Gemini experiences. Developers can access the model through Google AI Studio and the Gemini API, while enterprise customers can use it through Gemini Enterprise.Latest Gemini 3.8 UpdateThe launch comes only three weeks after Gemini 3.7 Flash and marks Google’s third Flash release within six weeks. Google says the new model improves software engineering, agentic tasks and multi-step reasoning while retaining Flash-level speed and cost efficiency. The company also highlights stronger prompt-injection robustness and safety measures for sensitive areas.Key HighlightsGemini 3.8 Flash launched on September 2, 2026.It targets coding, reasoning, agents and enterprise workflows.The model supports a context window of up to one million input tokens.Gemini 3.8 Flash Cyber focuses on defensive cybersecurity tasks.Cyber is available to selected trusted defenders through Fairwind.Introductory API pricing runs through December 31, 2026.ConclusionGemini 3.8 Flash expands Google’s Flash lineup with stronger reasoning, coding and agentic capabilities while keeping the model focused on speed and cost efficiency. The Cyber version takes the same underlying intelligence toward defensive cybersecurity and vulnerability remediation. Its restricted availability reflects Google’s effort to balance advanced cyber capabilities with safety controls.Disclaimer: This article is provided for general knowledge and news-information purposes only. Model capabilities, pricing, access, benchmarks and availability may change, so readers should verify the latest information through Google’s official channels.
