Research

Google Gemini 3.8 Flash Release Brief

Release brief for Gemini 3.8 Flash: September 2 2026 launch, $0.75/$3.75 introductory pricing through December 31, rollout to AI Mode and the Gemini app, the Flash Cyber variant, and brand-visibility implications.

By Ramanath, CTO & Co-Founder at Presenc AI · Last updated: September 2026

At a Glance

VendorGoogle
FamilyGemini 3 series
LaunchedGoogle released Gemini 3.8 Flash on September 2, 2026, three weeks after Gemini 3.7 Flash (August 13) and the third Flash update in three months. Google calls it its most intelligent workhorse model. A restricted sibling, Gemini 3.8 Flash Cyber, launched the same day for trusted security testers.
Context windowAbout 1,000,000 input tokens and 64,000 output tokens, with text, image, audio, video, and PDF input. These figures follow the Gemini 3.x Flash line and are reported consistently by independent trackers.
PricingIntroductory pricing of $0.75 per million input tokens and $3.75 per million output tokens through December 31, 2026, doubling to $1.50 and $7.50 from January 1, 2027. Context caching is $0.075 per million tokens during the introductory period. Batch and Flex tiers are half price; Priority costs 80 percent more. The price matches Gemini 3.7 Flash, so the upgrade costs nothing per token, though it can cost more per task.
Access channelsGemini API through Google AI Studio and Android Studio, Google Antigravity, Stitch, Gemini Enterprise, the Gemini app for Google AI Pro and Ultra subscribers, AI Mode in Google Search, and Gemini in Google Sheets. Gemini 3.8 Flash Cyber is available only through the new Fairwind Program for trusted defenders, government bodies, critical infrastructure operators, and software maintainers.

Notable Benchmarks

Google reports 54.9 percent on HLE-Verified and says 3.8 Flash beats 3.7 Flash on every benchmark it published, with strong results on DeepSWE v1.1, Vals Finance Agent V2, and Harvey's legal agent benchmark. For the Cyber variant, Google cites 47.2 percent pass@1 on CWE-Bench and says Chrome Security saw 2.6 times more correct patches than from competing commercial models.

Strengths

Frontier-adjacent coding and agentic performance at Flash pricing, broad multimodal input, and immediate distribution across Google's consumer surfaces including AI Mode.

Limitations

It is built on 3.7 Flash rather than a new base model, and Google says it deliberately does more work on complex tasks, running extra reasoning steps and repeated tool calls. That raises tokens per task. Google itself recommends staying on 3.7 Flash for efficiency-first workloads. The introductory price also ends on December 31, 2026.

Brand-Visibility Implications

The launch surface matters more than the benchmark sheet. Gemini 3.8 Flash went live in AI Mode on day one, so the model generating AI Mode answers changed for a very large search audience without any announcement inside Search itself. Its extra diligence, more reasoning steps and more tool calls, likely means more retrieval per answer, which gives well-structured, current pages more chances to be read and cited. Re-run AI Mode prompt sets after each Flash update rather than treating Google's AI surfaces as stable between I/O events. See Gemini 3.6 Flash brief, AI Mode defaulting to Gemini Flash, and Gemini brand visibility.

How Presenc AI Tracks This Model

Presenc AI monitors brand visibility on Google's Gemini 3 series as part of continuous multi-platform AI visibility tracking. We sample Google Gemini 3.8 Flash across representative prompt sets daily, compare against competitor performance on the same prompts, and flag material mention-rate changes so brand teams can respond quickly when AI representation shifts.

Frequently Asked Questions

Google released Gemini 3.8 Flash on September 2, 2026, three weeks after Gemini 3.7 Flash (August 13) and the third Flash update in three months. Google calls it its most intelligent workhorse model. A restricted sibling, Gemini 3.8 Flash Cyber, launched the same day for trusted security testers.
About 1,000,000 input tokens and 64,000 output tokens, with text, image, audio, video, and PDF input. These figures follow the Gemini 3.x Flash line and are reported consistently by independent trackers.
Gemini API through Google AI Studio and Android Studio, Google Antigravity, Stitch, Gemini Enterprise, the Gemini app for Google AI Pro and Ultra subscribers, AI Mode in Google Search, and Gemini in Google Sheets. Gemini 3.8 Flash Cyber is available only through the new Fairwind Program for trusted defenders, government bodies, critical infrastructure operators, and software maintainers.
The launch surface matters more than the benchmark sheet. Gemini 3.8 Flash went live in AI Mode on day one, so the model generating AI Mode answers changed for a very large search audience without any announcement inside Search itself. Its extra diligence, more reasoning steps and more tool calls, likely means more retrieval per answer, which gives well-structured, current pages more chances to be read and cited. Re-run AI Mode prompt sets after each Flash update rather than treating Google's AI surfaces as stable between I/O events. See Gemini 3.6 Flash brief, AI Mode defaulting to Gemini Flash, and Gemini brand visibility.

Track Your AI Visibility

See how your brand appears across ChatGPT, Claude, Perplexity, and other AI platforms. Start monitoring today.