Google has released Gemini 3.8 Flash, its third Flash-tier AI model in just six weeks, even as updates to its flagship Pro model remain stalled. The company describes the new release as its best reasoning and coding model to date, and it comes in two versions: a general-purpose "workhorse" model for agentic tasks and software development, and a specialised variant called Gemini 3.8 Flash Cyber, tuned for detecting and patching security vulnerabilities. The rapid pace of Flash releases, alongside the continued absence of a promised Gemini 3.5 Pro, suggests Google may be prioritising smaller, cheaper models while it struggles to match rivals on flagship performance.

Google is offering API access at an introductory rate of $0.75 per million input tokens and $3.75 per million output tokens until year's end, before prices rise to $1.50 and $7.50 respectively, a move likely driven by competitors cutting their own prices. Benchmark results show only marginal gains over the previous 3.7 Flash model overall, though coding performance improved notably, with Gemini 3.8 Flash topping the DeepSWE leaderboard for software engineering tasks. It still lags behind rivals such as Claude Opus on agentic computer-use tests, however. The Cyber variant reportedly achieved a 2.6x improvement in patch accuracy for Chrome's security team and found a critical vulnerability in two hours for Google Cloud, though it remains restricted to trusted testers and government users, while the standard model rolls out more broadly today.