Google has introduced Gemini 3.8, its latest reasoning and coding model, arriving as two variants: Gemini 3.8 Flash and Gemini 3.8 Flash Cyber. The company says 3.8 is its best reasoning and coding model yet, at the same speed and low cost as the previous version. This marks the third Flash release in six weeks, following 3.7 Flash from three weeks ago.
Gemini 3.8 Flash is positioned as Google's most intelligent workhorse model, with improvements over 3.7 Flash in software engineering, agentic tasks, and multi-step reasoning in specialized domains. It is priced at the same introductory rate as 3.7 Flash: $0.75 per million input tokens and $3.75 per million output tokens. For context, 3.7 Flash launched in August at that same introductory price, roughly half the cost of the earlier 3.6 Flash per million tokens.
Read more: Google Lets Gemini Users Hide Visible Watermarks on AI-Generated Media
The second variant, Gemini 3.8 Flash Cyber, is billed by Google as its most capable model for cybersecurity work, one the company says finds vulnerabilities and patches them automatically at frontier level. It is available to trusted defenders through the new Fairwind Program.
Both models share the same foundational intelligence, enhanced by long-running agentic loops that recursively evaluate and refine the underlying models. Google attributes the coding and reasoning gains to innovations including rigorous training in cybersecurity.
Gemini 3.8 Flash delivers substantial gains over 3.7 Flash, often approaching the performance of higher-cost frontier models, according to the announcement. The release follows Google's recent push into specialized AI offerings, including a financial services platform for Wall Street launched last month.
