Google Launches Gemini 4 Hydrogen

Google DeepMind SVP Koray Kavukcuoglu announced Gemini 4 Hydrogen, describing it as bringing advanced AI to a broad audience. The model is positioned specifically for agentic workflows—tasks where AI acts across multiple steps or tools with minimal human intervention. Key themes include multi-model efficiency, scalable developer and enterprise pipelines, and variable 'thinking levels' that let the model trade speed for depth depending on the task. Hydrogen is also integrated into consumer-facing products, including the Gemini app and Google Search. Google emphasizes safety and dependability as design requirements, not afterthoughts, particularly for large-scale deployments. The model is available to all users starting today. Full technical specs and benchmark data were not included in the announcement text, but the tiered intelligence system suggests a continuation of Google's strategy to serve both lightweight and compute-intensive use cases under one model family.