---
title: "Gemini 3.8 Flash Launches: Same-Price Upgrade for Long-Horizon Coding and Agents, with Flash Cyber Limited to Trusted Defenders"
url: https://xcube.enlightcorp.com.tw/en/news/google-gemini-3-8-flash-and-flash-cyber-long-horizon-agents
category: "Frontier Models"
source_type: official
published: 2026-09-02
updated: 2026-09-02
lang: en
---

# Gemini 3.8 Flash Launches: Same-Price Upgrade for Long-Horizon Coding and Agents, with Flash Cyber Limited to Trusted Defenders

Published 2026-09-02 · Updated 2026-09-02 · Official release · Source: [Google Blog](https://blog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber/)

**Key answer:** Gemini 3.8 Flash is a Flash-tier Google model released on September 2, 2026 for long-horizon software engineering and agent workflows, at an introductory $0.75 per million input and $3.75 per million output tokens. It matters because the high-volume Flash workhorse tier keeps getting more capable, and the price list already schedules an increase from January 1, 2027.

On September 2, 2026, Google announced Gemini 3.8 Flash and Gemini 3.8 Flash Cyber, in a post by Tulsee Doshi, Senior Director of Product Management, and Raluca Ada Popa, Gemini Security Lead at Google DeepMind. 3.8 Flash targets gains in long-horizon software engineering, multi-step problem solving, and agentic workflows at the same speed and cost as 3.7 Flash; Google describes it as a model that "works harder" through iterative reasoning and tool use on complex tasks.

3.8 Flash is available through Google AI Studio, the Gemini API, Android Studio, Stitch, and Gemini Enterprise, and is rolling into the Gemini app for Google AI Pro and Ultra subscribers, AI Mode in Google Search, and Google Sheets. Introductory pricing is $0.75 per million input tokens and $3.75 per million output tokens; starting January 1, 2027, it becomes $1.50 input and $7.50 output.

Gemini 3.8 Flash Cyber is a security-specialized variant with prioritized access through the new Fairwind Program for trusted government authorities, critical infrastructure operators, and software maintainers. Google cites its Chrome Security team: 3.8 Flash Cyber produced 2.6 times more correct patches to Chrome vulnerabilities than the best, much larger commercial models. Google also lists a 47.2% pass@1 on CWE-Bench patching.

Caveats: the benchmarks are vendor-reported, and the introductory price has a fixed end date, so budgets should be modeled on 2027 pricing. Separately, the Gemini API changelog announced on September 18 that access to Gemini 2.5 models is being limited to users who have actively used them, recommending 3.5 Flash-Lite or 3.8 Flash for new projects, which signals a faster retirement cadence for older models.

An enterprise considering Gemini 3.8 Flash behind its model gateway can build a checklist from the official information: confirm whether access goes through the Gemini API, Google AI Studio, or Gemini Enterprise, since billing and data-governance terms differ; model costs at both the introductory price through the end of 2026 and the $1.50 input / $7.50 output per million tokens that applies from January 1, 2027; plan to move any remaining Gemini 2.5 workloads to 3.5 Flash-Lite or 3.8 Flash per the September 18 changelog; and monitor tool-call counts and total tokens per task for long-horizon agent work, since Google says 3.8 Flash "works harder" through extra reasoning and tool steps.

What to watch next: tool-calling reliability and token consumption of 3.8 Flash on real agent tasks, its price-performance against other cloud workhorse models after the 2027 increase, and whether Fairwind-style restricted access for security models spreads to more products.

## X Cube view

With Flash-tier models updating every few weeks and older versions being gated, enterprises calling cloud models through a gateway should isolate concrete versions behind aliases or routing rules, rerun internal evals on each update, and model introductory prices with their expiry dates instead of hard-coding model IDs into applications.

Tags: Gemini, Gemini API, AI Agents, AI Security, Model Pricing, Google, Gemini 3.8 Flash, Gemini 3.8 Flash Cyber, Fairwind Program, Gemini Enterprise