Google releases Gemini 4 Argon, a new multimodal model that blends vision and text. The model improves on prior Gemini versions with higher accuracy on image‑captioning and reasoning tasks, and introduces a safety‑first training pipeline. Developers can access it via the Gemini API, enabling richer multimodal applications.
Opening Kapyn…