Google on September 2, 2026 introduced Gemini 3.8 Flash, the latest model in its Flash tier, at an introductory price of $0.75 per million input tokens and $3.75 per million output tokens through December 31, 2026, alongside a restricted cybersecurity variant called Gemini 3.8 Flash Cyber. The company’s announcement described the release as its third Flash update in six weeks, arriving three weeks after Gemini 3.7 Flash.
Google described Gemini 3.8 Flash as its most intelligent workhorse model, delivering significant improvements over 3.7 Flash across software engineering, agentic tasks, and multi-step reasoning in specialized domains. The model is generally available and ready for production use under the model ID gemini-3.8-flash.
Pricing and Availability
The introductory rate matches 3.7 Flash’s pricing and applies across Google AI Studio and the Gemini Enterprise Agent Platform. Starting January 1, 2027, standard pricing of $1.50 per million input tokens and $7.50 per million output tokens takes effect, according to the developer documentation.
Gemini 3.8 Flash supports a one-million-token context window, a maximum of 64,000 output tokens, and tunable thinking levels of low, medium, and high, with medium as the default. Google said the model takes smaller reasoning steps, calls tools iteratively, and verifies its work on difficult multi-step goals, and may use more tokens by design on complex tasks, especially at higher effort levels. For efficiency-first workloads, developers can lower the reasoning effort or continue using Gemini 3.7 Flash, which remains fully supported.
Developers can access the model through the Gemini API in Google AI Studio, Android Studio, and Google Antigravity, Google’s agent-first development environment. The Antigravity managed agent now runs on Gemini 3.8 Flash by default. Enterprises can reach it through Gemini Enterprise, and consumers with Google AI Pro or Ultra subscriptions can use it in the Gemini app, AI Mode in Google Search, and Gemini in Google Sheets.
Reported Performance
Google reported that Gemini 3.8 Flash often approaches the performance of higher-cost frontier models. On DeepSWE v1.1, a long-horizon software engineering benchmark, the company said the model outperforms most larger frontier models in autonomously solving complex engineering problems end to end. Google also reported that it outperforms 3.7 Flash and other frontier models on Vals Finance Agent V2 and Harvey’s Legal Agent Benchmark, and scores 54.9 percent on HLE-Verified, a benchmark covering multi-step reasoning across STEM, humanities, and professional fields.
Google attributed the gains to training in demanding domains, including cybersecurity, and to long-running agentic loops that recursively evaluate and refine the underlying models. Both new variants share the same foundational intelligence.
Gemini 3.8 Flash Cyber and the Fairwind Program
The second variant, Gemini 3.8 Flash Cyber, targets vulnerability detection and automated patching. Access is limited to trusted defenders through Google’s new Fairwind Program, which offers prioritized access to government authorities, critical infrastructure operators, and software maintainers who apply.
On CyberGym, an industry benchmark for autonomous vulnerability discovery, Google said the model reaches frontier-level performance, surpassing both 3.5 Flash Cyber and significantly larger frontier models. On an internal benchmark spanning 20 programming languages, Google reported a success rate exceeding 70 percent. On CWE-Bench, an external patching benchmark run by Collinear, the model recorded a pass@1 of 47.2 percent against a leading frontier model’s 47.8 percent, which Google characterized as Pareto-frontier performance at significantly lower cost.
Google said the Chrome Security team found the model produced 2.6 times more correct patches for Chrome vulnerabilities than the best larger commercial models. Wiz reported 7.5 to 9.7 percent higher recall on its internal penetration testing benchmark at 2.3 to 5.2 times lower cost than other leading frontier models, Google said. Google’s Cloud Vulnerability Research team used the model to find a critical foundational vulnerability in under two hours, a process the company said typically takes months.
Safety Measures
Google said Gemini 3.8 Flash ships with safeguards against misuse in chemical, biological, radiological, and nuclear domains and cyber offense under its Frontier Safety Framework. The Cyber variant carries a more permissive set of cybersecurity mitigations, which the company said is why it remains restricted to trusted defenders. Google also reported a significant improvement in prompt-injection robustness for the Gemini 3.8 models as measured by Gray Swan.
