Google shipped four Gemini Flash models in 106 days. But its flagship frontier model is still nowhere to be seen.

Google unveiled its latest AI model, Gemini 3.8 Flash, on Wednesday, boasting superior coding capabilities and competitive performance against rival AI models at a lower cost. The debut of Gemini 3.8 Flash follows three weeks after the release of its predecessor, Gemini 3.7 Flash, and marks Google's fourth Gemini Flash model since May.

Flash models are designed for users seeking low-cost responses without sacrificing much cognitive power. However, Google's flagship Gemini 3.5 Pro model, which CEO Sundar Pichai announced for June, is still in development. This has raised questions about Google's ability to catch up to the frontier AI of other technology companies.

Meta's chief AI officer, Alexander Wang, joked about Google's tardiness in the AI race following the launch of his company's new Muse Spark 1.3 model, which outperformed Google's models in a key benchmark. Google's Gemini 3.8 Flash ranks 10th on the Artificial Analysis Intelligence Index, a closely-watched performance benchmark.

Google has been releasing new Flash models rapidly, which some insiders believe showcases Google's progress in a key AI technique called recursive self-improvement (RSI). RSI involves an AI model that can improve and optimize its own code, creating even more capable versions of itself. While Google has been investing heavily in coding and reinforcement learning since the beginning of the year, its effort has not yet manifested in a more capable flagship model.