Google Photos adds AI video editing with cinematic relighting

Google launched Nano Banana 2 Lite, a lightweight image generation model built on Gemini 3.1 Flash-Lite architecture, capable of generating 1k resolution images in 4 seconds at $0.034 per 1,000 images. The model is available immediately to enterprise developers through Google AI Studio, the Gemini API, and GEAP. It trades resolution flexibility for speed and cost efficiency, targeting high-throughput commercial workflows like programmatic advertising and e-commerce asset generation.

Google imposed capacity limits on Meta's use of its Gemini AI models a few months ago, citing inability to meet the social media company's full demand. The restriction was not limited to Meta, as Google also constrained access for other clients. Google has since moved to address capacity issues by signing a deal to rent cloud computing capacity from Elon Musk's infrastructure.

Google is restructuring a months-old strike team focused on AI coding tools, aiming to improve model training and expand capabilities beyond coding into areas like presentation creation. The reorganization reflects competitive pressure from Anthropic and OpenAI, which are also broadening their AI coding tool applications. The changes also formalize what was originally conceived as a short-term group into a more permanent structure.

Google's DeepMind is investing approximately $75 million in A24, the independent film studio, to develop AI-powered movie production tools. This marks Google's first equity stake in a film studio and will span multiple projects focused on helping filmmakers expand their creative workflows. The non-exclusive collaboration aims to create tools shaped by filmmaker input rather than imposed from outside the industry.