Just three weeks after its previous model update, Google is speeding up its AI development cycle. The company officially launched Google Gemini 3.8 Flash alongside a dedicated security variant, Gemini 3.8 Flash Cyber. The release marks the third Flash update in just six weeks, bringing stronger reasoning and autonomous task execution while keeping the same budget-friendly pricing structure.
The core performance boost comes down to a simple design choice: the 3.8 Flash engine is built to work harder on complex requests. On difficult tasks, the model executes additional internal reasoning steps and calls tools iteratively to reach more accurate answers. Developers looking for lower compute overhead can adjust effort settings or stick with Gemini 3.7 Flash, which remains fully supported for efficiency-first workloads.
Enhanced long-horizon coding and agentic workflows
The standard Gemini 3.8 Flash model targets software engineering, agentic automation, and complex enterprise analysis. On the DeepSWE v1.1 benchmark for long-horizon software engineering, the model outperforms several larger frontier options in solving complex coding issues end-to-end at a fraction of the operating cost.
It also posts strong results across specialized knowledge fields. The model scored 54.9% on the HLE-Verified benchmark for multi-step reasoning across STEM and humanities. Plus, it tops previous iterations on professional evaluation tests like the Harvey Legal Agent Benchmark and Vals Finance Agent V2. Knowledge cutoff dates range from January 2025 to March 2026 depending on the specific domain.
Dedicated cybersecurity and automated patching
Alongside the general release, Google introduced Gemini 3.8 Flash Cyber, replacing its older 3.5 Cyber iteration. This one is available exclusively to vetted security researchers, government agencies, and critical infrastructure operators through Google’s new Fairwind Program. The model focuses on autonomous vulnerability discovery and automated code patching.
Internal testing demonstrates immediate practical impacts across Google’s own software stack:
- Chrome Security: Produced 2.6 times more accurate vulnerability patches in Chrome than significantly larger commercial alternatives.
- Penetration Testing: Delivered a 7.5% to 9.7% higher recall rate on internal Wiz benchmarks at a lower operational cost.
- Cloud Security: Helped Google’s Cloud Vulnerability Research team discover a critical foundational bug in under two hours. This process normally takes months.
Access options and promotional pricing
Google is maintaining its introductory pricing for Gemini 3.8 Flash through December 31. Rates are currently set at $0.75 per million input tokens and $3.75 per million output tokens.
Consumer access is already live for Google AI Pro and Ultra subscribers inside the Gemini app, AI Mode in Google Search, and Gemini in Google Sheets. Developers and enterprise clients can deploy the model immediately through Google AI Studio, the Gemini API, Android Studio, Stitch, and Gemini Enterprise.
The post Google Debuts Gemini 3.8 Flash and Cyber Variants with Massive Performance Leap appeared first on Android Headlines.
​Â
