Google Cloud launches updates for five models with different purposes

Google Cloud Launches Updates for Five Models with Different Purposes

Today at the Google Cloud Next conference, Google Cloud announced a series of model updates: The text-to-music model Lyria is now available for private preview on Vertex AI. By simply entering a text prompt, it can generate complete, production-ready materials; The text-to-video model Veo 2 has added editing and camera control functions, enabling precise optimization and reuse of video content; The voice generation model Chirp 3 now includes the Instant Custom Voice function, allowing users to create a custom voice by inputting just 10 seconds of audio; The text-to-image model Imagen 3 has improved its image generation and inpainting functions, which can reconstruct missing or damaged parts of an image and further enhance the quality of object removal editing. In addition, Google has launched Gemini 2.5 Flash, a major inference large model with low latency and high cost-effectiveness, which is now open for use.

—— Google Blog, Google Blog

via Windvane Reference Express - Telegram Channel