Google has released two new AI models, Gemini 3.8 Flash and Gemini 3.8 Flash Cyber, just three weeks after the previous generation. These models feature upgraded programming and reasoning capabilities, and introduce an automated cybersecurity remediation model.

Google announced the launch of two new models, Gemini 3.8 Flash and Gemini 3.8 Flash Cyber, on Wednesday, September 2, US ET, just three weeks after the release of the previous generation, Gemini 3.7 Flash. Gemini 3.8 Flash primarily targets long-cycle programming, agent workflows, and complex reasoning, with its high-reasoning tier achieving a comprehensive intelligence score of 59 in independent evaluations, a 3-point increase over 3.7 Flash. Gemini 3.8 Flash Cyber, on the other hand, focuses on vulnerability discovery and automated remediation in the cybersecurity domain, achieving a success rate of over 70% in vulnerability remediation during internal Google tests. Google stated that Gemini 3.8 is its "strongest reasoning and programming model" to date and will maintain the same low-price strategy as 3.7 Flash. However, independent evaluations show that the actual task cost for its high-reasoning tier has increased by approximately 40% due to increased token consumption. Additionally, Google simultaneously launched the Fairwind program, providing cybersecurity AI capabilities to over 650 government agencies, critical infrastructure, and large enterprise partners through restricted access.