AIBRIEF
Google's mid tier Gemini 3.8 Flash is its third Flash release in six weeks; the top end Gemini Pro line has not seen a new model since early 2026.
Google released Gemini 3.8 Flash on Tuesday, its third release in the cheaper, faster mid-tier "Flash" line in six weeks. The introductory API price is $0.75 per million input tokens and $3.75 per million output tokens through the end of 2026, with the regular rate set at $1.50 and $7.50 per million tokens (Ars Technica).
"Flash" is Google's mid-tier Gemini family: smaller and cheaper than the flagship Pro models, used by developers building chatbots, search tools, and coding assistants that trade raw capability for speed and cost. Google positioned 3.8 Flash as a routine upgrade, with only marginal gains over the July 3.7 Flash release on most benchmarks and more notable improvements on coding evaluations (Google blog).
A sibling variant, Gemini 3.8 Flash Cyber, ships tuned for vulnerability detection in code. Google's own evaluation PDF covers the Cyber model's security benchmark results; independent verification has not been published (evaluation PDF).
The release lands as AI labs broadly cut token prices to keep business customers from churning to rivals, per Ars Technica. The third-Flash-in-six-weeks pace, paired with no new Pro, is the part to watch over the next quarter.